Ebffiledoc 3513

Big Data Cloud and Applications Third
International Conference BDCA 2018

Kenitra Morocco April 4 5 2018 Revised
Selected Papers Youness Tabii
Visit to download the full and correct content document:
https://textbookfull.com/product/big-data-cloud-and-applications-third-international-co
nference-bdca-2018-kenitra-morocco-april-4-5-2018-revised-selected-papers-younes
s-tabii/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Intelligence Science and Big Data Engineering 8th

International Conference IScIDE 2018 Lanzhou China
August 18 19 2018 Revised Selected Papers Yuxin Peng
https://textbookfull.com/product/intelligence-science-and-big-
data-engineering-8th-international-conference-
iscide-2018-lanzhou-china-august-18-19-2018-revised-selected-
papers-yuxin-peng/
Advances in Computing and Data Sciences Second

International Conference ICACDS 2018 Dehradun India
April 20 21 2018 Revised Selected Papers Part I Mayank
Singh
https://textbookfull.com/product/advances-in-computing-and-data-
sciences-second-international-conference-icacds-2018-dehradun-
india-april-20-21-2018-revised-selected-papers-part-i-mayank-
singh/
Advances in Computing and Data Sciences Second

International Conference ICACDS 2018 Dehradun India
April 20 21 2018 Revised Selected Papers Part II Mayank
Singh
https://textbookfull.com/product/advances-in-computing-and-data-
sciences-second-international-conference-icacds-2018-dehradun-
india-april-20-21-2018-revised-selected-papers-part-ii-mayank-
singh/
Soft Computing Systems Second International Conference

ICSCS 2018 Kollam India April 19 20 2018 Revised
Selected Papers Ivan Zelinka
https://textbookfull.com/product/soft-computing-systems-second-
international-conference-icscs-2018-kollam-india-
april-19-20-2018-revised-selected-papers-ivan-zelinka/
Parallel Computational Technologies 12th International
Conference PCT 2018 Rostov on Don Russia April 2 6 2018
Revised Selected Papers Leonid Sokolinsky
https://textbookfull.com/product/parallel-computational-
technologies-12th-international-conference-pct-2018-rostov-on-
don-russia-april-2-6-2018-revised-selected-papers-leonid-
sokolinsky/
Cloud Computing and Security 4th International

Conference ICCCS 2018 Haikou China June 8 10 2018
Revised Selected Papers Part I Xingming Sun
https://textbookfull.com/product/cloud-computing-and-
security-4th-international-conference-icccs-2018-haikou-china-
june-8-10-2018-revised-selected-papers-part-i-xingming-sun/

Revised Selected Papers Part II Xingming Sun
june-8-10-2018-revised-selected-papers-part-ii-xingming-sun/

Revised Selected Papers Part VI Xingming Sun
june-8-10-2018-revised-selected-papers-part-vi-xingming-sun/
Cyber Physical Systems Model Based Design 8th

International Workshop CyPhy 2018 and 14th
International Workshop WESE 2018 Turin Italy October 4
5 2018 Revised Selected Papers Roger Chamberlain
https://textbookfull.com/product/cyber-physical-systems-model-
based-design-8th-international-workshop-cyphy-2018-and-14th-
international-workshop-wese-2018-turin-italy-
Youness Tabii
Mohamed Lazaar
Mohammed Al Achhab
Nourddine Enneya (Eds.)
Communications in Computer and Information Science 872
Big Data, Cloud

and Applications
Third International Conference, BDCA 2018
Kenitra, Morocco, April 4–5, 2018
Revised Selected Papers
123
Communications
in Computer and Information Science 872
Commenced Publication in 2007
Founding and Former Series Editors:
Phoebe Chen, Alfredo Cuzzocrea, Xiaoyong Du, Orhun Kara, Ting Liu,
Dominik Ślęzak, and Xiaokang Yang
Editorial Board
Simone Diniz Junqueira Barbosa
Pontifical Catholic University of Rio de Janeiro (PUC-Rio),
Rio de Janeiro, Brazil
Joaquim Filipe
Polytechnic Institute of Setúbal, Setúbal, Portugal
Igor Kotenko
St. Petersburg Institute for Informatics and Automation of the Russian
Academy of Sciences, St. Petersburg, Russia
Krishna M. Sivalingam
Indian Institute of Technology Madras, Chennai, India
Takashi Washio
Osaka University, Osaka, Japan
Junsong Yuan
University at Buffalo, The State University of New York, Buffalo, USA
Lizhu Zhou
Tsinghua University, Beijing, China
More information about this series at http://www.springer.com/series/7899
Youness Tabii Mohamed Lazaar
•
Mohammed Al Achhab Nourddine Enneya (Eds.)

•
Big Data, Cloud

and Applications
Third International Conference, BDCA 2018
Kenitra, Morocco, April 4–5, 2018
Revised Selected Papers
123
Editors
Youness Tabii Mohammed Al Achhab
Abdelmalek Essaâdi University Abdelmalek Essaâdi University
Tétouan Tétouan
Morocco Morocco
Mohamed Lazaar Nourddine Enneya
Abdelmalek Essaâdi University Université Ibn-Tofail
Tétouan Tétouan
Morocco Morocco
ISSN 1865-0929 ISSN 1865-0937 (electronic)

Communications in Computer and Information Science
ISBN 978-3-319-96291-7 ISBN 978-3-319-96292-4 (eBook)
https://doi.org/10.1007/978-3-319-96292-4
Library of Congress Control Number: 2018948223
© Springer Nature Switzerland AG 2018

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the
material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now
known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book are
believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors
give a warranty, express or implied, with respect to the material contained herein or for any errors or
omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
We are happy to present you this book, Big Data, Cloud and Applications, which is a
collection of papers that were presented at the 3rd International Conference on Big
Data Cloud and Applications, BDCA 2018. The conference took place on April 04–05,
2018, in Kenitra, Morocco.
The book consisted of nine chapters, which correspond to the four major areas that
are covered during the conference, namely, Big Data, Cloud Computing, Maching
Learning, Deep Learning, Data Analysis, Neural Networks, Information System and
Social Media, Natural Language Processing, Image Processing and Applications.
Every year BDCA attracted researchers from all over the world, and this year was
not an exception – we received 99 submissions from 12 countries. More importantly,
there were participants from many countries, which indicates that the conference is
truly gaining more and more international recognition as it brought together a vast
number of specialists who represented the aforementioned fields and share information
about their newest projects. Since we strived to make the conference presentations and
proceedings of the highest quality possible, we only accepted papers that presented the
results of various investigations directed to the discovery of new scientific knowledge
in the area of Big Data, Cloud Computing and their applications. Hence, only 45 papers
were accepted for publishing (i.e., 45% acceptance rate). All the papers were reviewed
and selected by the Program Committee, which comprised 96 reviewers from over 58
academic institutions. As usual, each submission was reviewed following a double
process by at least two reviewers. When necessary, some of the papers were reviewed
by three or four reviewers. Our deepest thanks and appreciation go to all the reviewers
for devoting their precious time to produce truly through reviews and feedback to the
authors.
July 2018 Youness Tabii

Mohamed Lazaar
Mohammed Al Achhab
Nourddine Enneya
Organization
The 3rd International Conference on Big Data, Cloud and Applications (BDCA 2018)
was organized by Abdelmalek Essaadi University and IbnTofail University and was in
Kenitra, Morocco (April 04–05, 2018).
General Chairs
Youness Tabii National School of Applied Sciences (ENSA), Tetouan,
Morocco
Nourddine Enneya Faculty of Sciences, Kenitra, Morocco
Local Organizing Committee

Nourddine Enneya FS, Ibn Tofail University, Kenitra, Morocco
Jihane Alami Chentoufi FS, Ibn Tofail University, Kenitra, Morocco
Jalal Laassiri FS, Ibn Tofail University, Kenitra, Morocco
Abdelalim Sadiq FS, Ibn Tofail University, Kenitra, Morocco
Youness Tabii ENSA, Abdelmalek Essaadi University, Tetouan,
Morocco
Mohamed Lazaar ENSA, Abdelmalek Essaadi University, Tetouan,
Morocco
Mohamed Al Achhab ENSA, Abdelmalek Essaadi University, Tetouan,
Morocco
Mohamed Chrayah ENSA, Abdelmalek Essaadi University, Tetouan,
Morocco
Btissam Dkhissi ENSA, Abdelmalek Essaadi University, Tetouan,
Morocco
Program Committee
Hamid R. Arabnia University of Georgia, USA
Abdelkaher Ait Ibn Zohr University, Morocco
Abdelouahad
Noura Aknin FS, Abdelmalek Essaadi University, Morocco
Adel Alimi REGIM, Sfax University, Tunisia
Mohammed Al Achhab ENSA, Abdelmalek Essaadi University, Morocco
Naoual Attaoui FS, Abdelmalek Essaadi University, Morocco
Abderrahim Azouani Mohammed 1st University, Morocco
Jenny Benois-Pineau Bordeaux University, France
Abdellah Abouabdellah ENSA, Ibn Tofail University, Morocco
Amel Benazza Supcom Carthage University, Tunisia
VIII Organization
Kamal Baraka Cadi Ayyad University, Morocco

Mohamed Batouche Constantine University 2, Algeria
Lamia Benameur FS, Abdelmalek Essaadi University, Morocco
Hamid Bennis EST, Moulay Ismail University, Morocco
Mohamed Ben Halima REGIM, Sfax University, Tunisia
Fadila Bentayeb Lyon 2 University, France
Samir Bennani EMI, Mohammed V University, Morocco
Thierry Berger Limoges University, France
Kamel Besbes FSM, University of Monastir, Tunisia
Mustapha Boushaba Montréal University, Canada
Aoued Boukelif University of Sidi-Bel-Abbès, Algeria
Abdelhak Boulaalam FP, Sidi Mohamed Ben Abdellah University, Morocco
Abdelhani Boukrouche Guelma University, Algeria
Jaouad Boukachour ISEL le Havre, France
Omar Boussaid Lyon 2 University, France
Anne Canteaut Inria-Rocquencourt, France
Claude Carlet Paris 8 University, France
Mohamed Chrayah ENSA, Abdelmalek Essaadi University, Morocco
Habiba Chaoui ENSA, Ibn Tofail University, Morocco
Btissam Dkhissi ENSA, Abdelmalek Essaadi University, Morocco
Abdellatif El Afia ENSIAS, Mohammed V University, Morocco
Nabil El Akkad ENSA, Hassan 1st University, Morocco
Youssouf El Allioui Hassan 1st University, Morocco
Younès El Bouzekri El ENSA, Ibn Tofail University, Morocco
Idrissi
Abdelaziz El Hibaoui FS, Abdelmalek Essaadi University, Morocco
Mohammed Elghzaoui FP, University Mohammed 1st, Morocco
Kamal Eddine El Kadiri ENSA, University of Abdelmalek Essaadi, Morocco
Said El Kafhali Hassan 1st University, Morocco
Yasser Elmadani Elalami Sidi Mohamed Ben Abdellah University, Morocco
Abderrahim El Mhouti FST, Mohammed 1st University, Morocco
Mourad El Yadari FP, Moulay Ismail University, Morocco
El Mokhtar En-Naimi FST, Abdelmalek Essaadi University, Morocco
Noureddine Ennahnahi Sidi Mohamed Ben Abdellah University, Morocco
Karim El Moutaouakil ENSA, Mohammed 1st University, Morocco
Nourddine Enneya Faculty of Sciences, Kenitra, Morocco
Abdelkarim Erradi Qatar University, Doha, Qatar
Mohamed Ettaouil FST, Sidi Mohamed Ben Abdellah University,
Morocco
Siti Zaiton Mohd Hashim University Teknologi, Malaysia
Adel Hafiane INSA Centre Val de Loire, France
Abdelhakim Hafid Montréal University, Canada
Abderrahmane Habbal Inria Sophia Antipolis, France
Faïez Gargouri University of Sfax, Tunisia
Youssef Ghanou EST, Moula Ismail University, Morocco
Khalid Haddouch ENSA, Mohammed 1st University, Morocco
Organization IX
Ebroul Izquierdo Queen Mary, University of London, UK

Mohamed Hanini Hassan 1st University, Morocco
Yanguo Jing London Metropolitan University, UK
Ismail Jellouli FS, Abdelmalek Essaadi University, Morocco
Joel J. P. C. Rodrigues Beira Interior University, Portugal
Asiya Khan Plymouth University, UK
Mejdi Kaddour Oran University, Algeria
Eleni Karatza Aristotle University of Thessaloniki, Greece
Hichem Karray REGIM, Sfax University, Tunisia
Epaminondas Kapetanios FST, WU, London, UK
Driss Laanaoui Cadi Ayyad University, Morocco
Tarik Lamoudan University of King Khalid, Abha, KSA
Yacine Lafifi Guelma University, Algeria
Mohamed Lazaar ENSA, Abdelmalek Essaadi University, Morocco
Mark Leeson School of Engineering, University of Warwick, UK
Pascal Lorenz University of Haute Alsace, France
Chakir Loqman FS, Sidi Mohamed Ben Abdellah University, Morocco
Lin Ma Huawei Noah’s Ark Lab, Hong Kong, China
Mostafa Merras Sidi Mohamed Ben Abdellah University, Morocco
Souham Meshoul University Constantine 2, Algeria
Abdellatif Medouri ENSA, Abdelmalek Essaadi University, Morocco
Safia Nait-Bahloul Oran University, Algeria
Nidal Nasser Alfaisal University, KSA
Rachid Oulad Haj Thami ENSIAS, Mohammed V University, Morocco
Barbaros Preveze Çankaya University, Turkey
Gabriella Sanniti Di Baja ICAR-CNR, Naples, Italy
Abdelalim Sadiq FS, Ibn Tofail University, Morocco
Chafik Samir University of Clermont Auvergne, France
M’hamed Ait Kbir FST, Abdelmalek Essaadi University, Morocco
Khaled Salah Khalifa University, Abu Dhabi, UAE
Hassan Satori Mohammed 1st University, Morocco
Patrick Siarry Paris-Est Créteil University, France
Hassan Silkan FS, Chouaib Doukkali University, Morocco
Sahbi Sidhom Lorraine University, Nancy, France
Mohammad Stanford University, USA
Shokoohi-Yekta
Youness Tabii ENSA, Abdelmalek Essaadi University, Morocco
Nawel Takouachet ESTIA Technopole Izarbel – France
Jamal Zbitou Hassan 1st University, Morocco
Abdelhamid Zouhair ENSA, Mohammed 1st University, Morocco
Ali Wali REGIM Sfax University, Tunisia
Said Elhajji Mohammed V University, Rabat, Morocco
Contents
Big Data
Informal Learning in Twitter: Architecture of Data Analysis Workflow

and Extraction of Top Group of Connected Hashtags. . . . . . . . . . . . . . . . . . 3
Abdelmajid Chaffai, Larbi Hassouni, and Houda Anoun
A MapReduce-Based Adjoint Method to Predict the Levenson Self Report

Psychopathy Scale Value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Manal Zettam, Jalal Laassiri, and Nourdddine Enneya
Big Data Optimisation Among RDDs Persistence in Apache Spark . . . . . . . . 29

Khadija Aziz, Dounia Zaidouni, and Mostafa Bellafkih
Cloud Computing
QoS in the Cloud Computing: A Load Balancing Approach

Using Simulated Annealing Algorithm. . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Mohamed Hanine and El Habib Benlahmar
A Proposed Approach to Reduce the Vulnerability in a Cloud System . . . . . . 55

Chaimae Saadi and Habiba Chaoui
A Multi-factor Authentication Scheme to Strength Data-Storage Access . . . . . 67

Soufiane Sail and Halima Bouden
A Novel Text Encryption Algorithm Based on the Two-Square

Cipher and Caesar Cipher . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Mohammed Es-Sabry, Nabil El Akkad, Mostafa Merras,
Abderrahim Saaidi, and Khalid Satori
Machine Learning
Improving Sentiment Analysis of Moroccan Tweets

Using Ensemble Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Ahmed Oussous, Ayoub Ait Lahcen, and Samir Belfkih
Comparative Study of Feature Engineering Techniques

for Disease Prediction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Khandaker Tasnim Huq, Abdus Selim Mollah,
and Md. Shakhawat Hossain Sajal
XII Contents
Business Process Instances Scheduling with Human Resources

Based on Event Priority Determination . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
Abir Ismaili-Alaoui, Khalid Benali, Karim Baïna,
and Jamal Baïna
Hashtag Recommendation Using Word Sequences’ Embeddings . . . . . . . . . . 131

Nada Ben-Lhachemi and El Habib Nfaoui
Towards for Using Spectral Clustering in Graph Mining . . . . . . . . . . . . . . . 144

Z. Ait El Mouden, R. Moulay Taj, A. Jakimi, and M. Hajar
Automatic Classification of Air Pollution and Human Health . . . . . . . . . . . . 160

Rachida El Morabet, Abderrahmane Adoui El Ouadrhiri,
Jaroslav Burian, Said Jai Andaloussi, Said El Mouak,
and Abderrahim Sekkaki
Deep Learning
Deep Semi-supervised Learning for Virtual Screening Based

on Big Data Analytics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
Meriem Bahi and Mohamed Batouche
Using Deep Learning Word Embeddings for Citations Similarity

in Academic Papers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
Oumaima Hourrane, Sara Mifrah, El Habib Benlahmar,
Nadia Bouhriz, and Mohamed Rachdi
Using Unsupervised Machine Learning for Data Quality.

Application to Financial Governmental Data Integration. . . . . . . . . . . . . . . . 197
Hanae Necba, Maryem Rhanoui, and Bouchra El Asri
Advanced Machine Learning Models for Large Scale Gene Expression

Analysis in Cancer Classification: Deep Learning Versus
Classical Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
Imene Zenbout and Souham Meshoul
Stemming and Lemmatization for Information Retrieval Systems

in Amazigh Language . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
Amri Samir and Zenkouar Lahbib
Data Analysis
Splitting Method for Decision Tree Based on Similarity with Mixed Fuzzy
Categorical and Numeric Attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
Houda Zaim, Mohammed Ramdani, and Adil Haddi
Contents XIII
Mobility of Web of Things: A Distributed Semantic

Discovery Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
Ismail Nadim, Yassine El Ghayam, and Abdelalim Sadiq
Comparison of Feature Selection Methods for Sentiment Analysis. . . . . . . . . 261

Soufiane El Mrabti, Mohammed Al Achhab, and Mohamed Lazaar
A Hierarchical Nonlinear Discriminant Classifier Trained

Through an Evolutionary Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
Ziauddin Ursani and David W. Corne
A Feature Level Fusion Scheme for Robust Speaker Identification . . . . . . . . 289

Sara Sekkate, Mohammed Khalil, and Abdellah Adib
One Class Genetic-Based Feature Selection for Classification

in Large Datasets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
Murad Alkubabji, Mohammed Aldasht, and Safa Adi
Multiobjective Local Search Based Hybrid Algorithm for Vehicle

Routing Problem with Soft Time Windows . . . . . . . . . . . . . . . . . . . . . . . . 312
Bouziyane Bouchra, Dkhissi Btissam, and Cherkaoui Mohammad
Dimension Reduction Techniques for Signal Separation Algorithms . . . . . . . 326

Houda Abouzid and Otman Chakkor
Neural Networks
A Probabilistic Vector Representation and Neural Network

for Text Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343
Mariem Bounabi, Karim El Moutaouakil, and Khalid Satori
Improving Implementation of Keystroke Dynamics Using K-NN

and Manhattan Distance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356
Farida Jaha and Ali Kartit
SARIMA Model of Bioelectic Potential Dataset . . . . . . . . . . . . . . . . . . . . . 367

Imam Tahyudin, Berlilana, and Hidetaka Nambo
New Starting Point of the Continuous Hopfield Network . . . . . . . . . . . . . . . 379

Khalid Haddouch and Karim El Moutaouakil
Information System And Social Media
A Concise Survey on Content Recommendations . . . . . . . . . . . . . . . . . . . . 393

Mehdi Srifi, Badr Ait Hammou, Ayoub Ait Lahcen,
and Salma Mouline
XIV Contents
Toward a Model of Agility and Business IT Alignment . . . . . . . . . . . . . . . . 406

Kawtar Imgharene, Karim Doumi, and Salah Baina
Integration of Heterogeneous Classical Data Sources

in an Ontological Database. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417
Oussama El Hajjamy, Larbi Alaoui, and Mohamed Bahaj
Toward a Solution to Interoperability and Portability of Content

Between Different Content Management System (CMS): Introduction
to DB2EAV API. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433
Abdelkader Rhouati, Jamal Berrich, Mohammed Ghaouth Belkasmi,
and Toumi Bouchentouf
Image Processing and Applications
Reconstruction of the 3D Scenes from the Matching Between Image Pair

Taken by an Uncalibrated Camera. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 447
Karima Karim, Nabil El Akkad, and Khalid Satori
An Enhanced MSER Based Method for Detecting Text in License Plates. . . . 464
Mohamed Admi, Sanaa El Fkihi, and Rdouan Faizi
Similarity Performance of Keyframes Extraction on Bounded Content

of Motion Histogram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 475
Abderrahmane Adoui El Ouadrhiri, Said Jai Andaloussi,
El Mehdi Saoudi, Ouail Ouchetto, and Abderrahim Sekkaki
Natural Language Processing
Modeling and Development of the Linguistic Knowledge Base DELSOM . . . 489

Fadoua Mansouri, Sadiq Abdelalim, and Youness Tabii
Incorporation of Linguistic Features in Machine Translation Evaluation

of Arabic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 500
Mohamed El Marouani, Tarik Boudaa, and Nourddine Enneya
Effect of the Sub-graphemes’ Size on the Performance of Off-Line

Arabic Writer Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 512
Nabil Bendaoud, Yaâcoub Hannad, Abdelillah Samaa,
and Mohamed El Youssfi El Kettani
Arabic Text Generation Using Recurrent Neural Networks . . . . . . . . . . . . . . 523

Adnan Souri, Zakaria El Maazouzi, Mohammed Al Achhab,
and Badr Eddine El Mohajir
Contents XV
Integrating Corpus-Based Analyses in Language Teaching and Learning:

Challenges and Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 534
Imad Zeroual, Anoual El Kah, and Abdelhak Lakhouaja
Arabic Temporal Expression Tagging and Normalization . . . . . . . . . . . . . . . 546

Tarik Boudaa, Mohamed El Marouani, and Nourddine Enneya
Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 559

Big Data
Informal Learning in Twitter: Architecture
of Data Analysis Workflow and Extraction
of Top Group of Connected Hashtags
Abdelmajid Chaffai ✉ , Larbi Hassouni, and Houda Anoun

( )
RITM LAB, CED Engineering Sciences, ENSEM, Hassan II University Casablanca,

Casablanca, Morocco
majedchaffai@gmail.com
Abstract. The Advance of web-based technologies have brought radical

changes to web site design and web service usage, primarily in terms of interactive
contents and user engagement in collaboration and information sharing. In
nutshell the web has been transformed from static media to the preferred commu‐
nication media where the user is a key player in the creation of his experiences.
The increase in the popularity of social networks on the Web has shaken up tradi‐
tional models in different areas, including learning. Many individuals have
resorted to social networking to educate themselves. Such learning is close to
natural learning, the learner is autonomous to draw the pathway which best suits
his individual needs in order to upgrade his skills. Several training organizations
use the Twitter platform to announce the training they provide. We conduct an
experiment on twitters data which are related to the training themes in Big Data
and Data Science, we perform an exploratory analysis and extract the top group
of connected hashtags using the Graph X library provided by the Spark frame‐
work. Data that come from the Twitter platform is produced at high speed and in
a complex structure. This leads us to use a distributed infrastructure based on two
efficient frameworks Apache Hadoop and Spark. Data ingestion layer is built by
combining two frameworks Apache Flume and Kafka.
Keywords: Informal learning · Social network data · Distributed environment

Apache spark · Graph · Connected components
1 Introduction
The learning is a long life process which takes place everywhere; it is divided in two
categories [1] formal and non-formal or informal. Formal learning is often validated by
official certifications; education occurs in structured environments such as schools and
universities and is supervised by teachers. Knowledge and skills acquired outside the
formal setting enable an informal learning. In today’s world, communication between
people occurs often through the use of social media platforms, wikis, micro-blogs which
become the main channels for conveying and sharing information in a quickly manner.
Communities and groups have been built around common points of interest. With
advances in Web2 technologies, the user of social network platforms once authenticated,
© Springer Nature Switzerland AG 2018

Y. Tabii et al. (Eds.): BDCA 2018, CCIS 872, pp. 3–15, 2018.
https://doi.org/10.1007/978-3-319-96292-4_1
4 A. Chaffai et al.
he can freely have several roles, read other people’s posts, write messages, insert media
and documents, search people and trend topics. Although social networks are considered
as entertainment spaces, several universities are attracted by the insertion of informal
learning via social networks like Twitter in their academic development [2]. In fact, in
this new age of data and computing, many individuals, students in higher education or
professionals have resorted to informal means to educate themselves and upgrade their
skills for example in cutting edge of tools in information technologies by working on-
line short courses and workshops. Informal learning through social media leads to
empowerment and self-efficacy while saving time and money in the learning process
and increase visibility in society.
Social network Analytics [3] is a set of methods and technologies that allow
collecting a large datasets from social network platforms sources, transform them in a
way that they become available and ready to be consumed by analysts. Text mining,
Natural language processing, classification and clustering algorithms are used to extract
the hidden insights in order to improve the best knowing of the user’s experiences. New
open source technologies like Apache Hadoop [4] and Spark [5] allow building infra‐
structures which aimed to manage massive datasets by distributing storage and
computing across clusters of low cost machines, they handle and combine both struc‐
tured and unstructured data that come from internal and external data sources.
Depending on data production, data processing tasks is divided into two groups:
– Batch Processing: data are collected in big batches over period of time, it is stored
in distributed file system, then processing and analysis jobs are applied at once, and
batch results are generated.
– Streaming Processing: data come in continuous way; processing and analysis jobs
are applied in near real time or in small time. In this work we use Apache Spark as
data processing engine, it is a distributed framework developed in Scala programming
language and works as a Java Virtual Machine, Spark is designed for fast scalable
in-memory computing and relies on Hadoop to run in cluster mode and use HDFS
[6] storage, it comes with a high level programming model that hides the partitioning
of dataset in memory of cluster, using a novel data structure called Resilient Distrib‐
uted Dataset RDD [7] which is an immutable distributed collection of objects parti‐
tioned across different nodes of the cluster. RDD data-sharing abstraction allows to
use wide range of APIs provided by Spark: Spark SQL, Spark Streaming, MLlib
(Machine Learning library), and GraphX (graph processing). Apache spark is suited
to perform analytics that need iterative operations. It allows to process data directly,
comparing to Map Reduce [8] programs which need several access to disk to retrieve
intermediate result. Since twitter data are generated at high speed and in a complex
structure, we implement a hybrid architecture which provides a faster ETL based on
data pipeline that ensures the data collection and processing in a unified and distrib‐
uted environment. We have conducted an experiment on twitters data filtered by
keywords associated with 6 topics of big data technologies and data science which
are of hot interests to developer and industrial communities.
In this paper we describe the necessary steps to carry out an exploratory analysis and
the extraction of the top group of connected hashtags.
Informal Learning in Twitter 5
The rest of the paper is structured as follows. Section 2 discusses related work.
Section 3 describes the Architecture of Data analysis workflow, Sect. 4 discusses the
experiment and finally, Sect. 5 concludes the paper.
2 Related Work
Social network analysis is an emerging research field which aims to better understand
how people seek and share information in social network platforms. Bonchi et al. in [9]
provided an overview of what we consider to be key problems and techniques in social
network analysis from a business applications perspective. The authors described each
area of research in the context of a specific business processes classification framework
(The APQC process classification framework), and then focused on several areas, giving
an overview of the main problems and describing state-of-the-art approaches.
The explosion of the use of micro blogs by students offers opportunities to exploit
this new communication channel in process-oriented learning. In paper [10], the authors
proposed a platform which uses Twitter news in Education known as NIE in order to
provide the latest news classified on various topics then enable discussion and debate
groups. They implemented a prototype system which uses Twitter as source to the hot
news and trends. For classification topics, each news tweet is cleaned and mapped into
its words. The Naïve Bayes classifier is used to achieve the classification based on
predefined number of keywords which correspond to the selected topics. The platform
offers to learners a News Visualizer using treemap to facilitate the learner’s query which
is based on period, keywords, and desired topic. Cosine similarity method, Based on
user document similarity and hierarchical agglomerative clustering is used to study the
learners’ preferences.
Aramo-Immonen et al. [11] employ Twitter data to study interactions between
members of community of managers attending a conference. Data are retrieved two
weeks before the conference. The process of data-driven visual network analytics and
the Ostinato [12] process model are provided to extract insights into the informal
learning of community managers. Quantitative and qualitative analyses of Twitter data
are produced like analysis of the top hash tags over time before the conference and the
network of hash tag co-occurrences.
In paper [13], the authors developed a workflow that consists to integrate both qual‐
itative analysis and large-scale data mining techniques. They focused on engineering
students’ Twitter posts to understand issues and problems in their educational experi‐
ences. The authors conducted a qualitative analysis on samples taken from about 25,000
tweets related to engineering students’ college life. They found engineering students
encounter problems such as heavy study load, lack of social engagement, and sleep
deprivation. A multi-label classification algorithm is implemented to classify tweets
reflecting students’ problems.
The majority of tweets do not contain the geographical location through exact GPS
coordinates (latitude and longitude). The authors attempt in [14] to identify a location
of the tweets. They employ twitter data to fit a Naive bayes model in order to classify a
tweets based on features as users’ timezone, the user’s language, and the parsed users’
location. The classifier with an accuracy of 82% was achieved and performs well on
active Twitter countries such as the Netherlands and United Kingdom.
An analysis of errors made by the classifier shows that mistakes were made due to
limited information and shared properties between countries such as shared timezone.
A feature analysis was performed in order to see the effect of different features. The
features timezone and parsed user location were the most informative features.
3 Twitter Data Characteristics and Architecture of Data Analysis

Workflow
Twitter has become a largest social space in the world where 330 million monthly active
users, discuss several topics and publish 500 million tweets per day. This data source
offers tremendous opportunities to analyze social trends for multiple purposes. Twitter
offers two types of APIs, Rest API and streaming APIs (for developers in real time) that
allow different clients applications written in different languages [15] to consume the
tweets. For example, in case of Java and Scala, Twitter4J is an open source Java library
used for interfacing with Twitter’s Application Programming Interfaces (APIs). Tweets
data come in non-structured nature, they are encoded using Java Script Object Notation
(JSON) based on key-value pairs. Each tweet has an author (user), a message, a unique
ID, a timestamp of when it was created, and geo metadata often turned off by users. Each
User has a Twitter name, an ID, a number of followers. Tweet contains ‘entity’ objects,
which are arrays of contents such as hashtags, mentions, media, and links.
A typical SNA workflow consists of several interacting phases which are:
• Data collection
• Data preparation
• Data analysis
• Insights.
The different topics discussed in the context of informal learning and social learning
in twitter are very varied, in this paper we propose a flexible data system (see Fig. 1)
capable to receive data from different topics through multiple agents, each agent inter‐
cepts the stream data in real time based on keywords related to a given topic, Apache
Flume [16] is used in the data collection layer. We are faced with a case where there
will be several flume-agents, so we need a strategy to categorize the message, for this
we use Apache Kafka [17] as an efficient publish-subscribe messaging system to separate
the incoming data in topics and keep them in scalable and fault-tolerant way. In the rest
of data pipeline, we use Spark streaming to consume, parse the incoming data in real
time and store them in HDFS storage. Analysis tasks to extract insights can be performed
by using Spark SQL and Spark ML.
Fig. 1. Overall architecture of the proposed SNA workflow.
4 Experiment
4.1 General Description
Due to strong competition between organizations for integrating data into decision
making, hiring opportunities for data specialists and data infrastructure specialists are
much greater than those of other profiles. We will study this trend in the twitter social
network as a case study, to try to extract useful information about users who are inter‐
ested in acquiring new knowledge or who share their experiences in the field of big data.
We employ data from twitter that is filtered based on the following keywords: “bigdata”,
“datascience”, “machineLearning”, “hadoop”, “spark”, “analytics”.
4.2 Environment Experiment
We deployed a small local cluster for Hadoop and Spark on 11 nodes running Ubuntu
16.04 LTS and interconnected via one switch of 1 Gb/s. The Hadoop cluster is built
using Hadoop version 2.7.3. The Spark cluster is built using Spark version 2.0.0. One
machine is designed as Master for both Spark and Hadoop, the others nodes are both
the Hadoop slaves and Spark workers. The following configuration is the same for all
nodes: Intel(R) Core(TM) i5-3470 CPU 3.20 GHz(4CPUs), 1 Gb/s network connection,
300 GB hard disk, 8 GB Memory.
4.3 Methodology
Data Ingestion
Retrieving data from the Twitter API requires credentials that can be obtained from
https://apps.twitter.com/, we register our application as a twitter app, then the authori‐
zation parameters are generated as follows: Consumer Key (API Key), Consumer Secret
(API Secret), Access Token and Access Token Secret.
Apache Flume is used to collect tweets data in JSON format from the source and
move it to Kafka in plaintext. As defined on its site [18], “Flume is a distributed and
available service for efficiently collecting, aggregating, and moving large amounts of
log data. It has a simple and flexible architecture based on streaming data flows.” The
main components of flume data pipeline (see Fig. 2) are source, channel, and sinks.
Flume agent is a JVM daemon responsible to manage the data flow. The source contin‐
uously retrieves tweets data in JSON format based on several keywords from the Twitter.
The channel act as a passive storage, it maintains the event data until a next hop which
is a Kafka cluster.
Fig. 2. Flume architecture. Fig. 3. Kafka concept.
The main components of Kafka-based architecture are shown in Fig. 3:

• Broker: Kafka is a cluster of nodes, each a node is a broker.
• Topic: is a category of related messages.
• Producer: each application that produces and sends the messages to Kafka topic for
example our flume-agent.
• Consumer: each application that subscribes to kafka topic and consumes the
messages.
Kafka relies on Zookeeper to manage his components and for monitoring the status
of operations that occur on the cluster.
We create one topic with 3 replicated partitions as shown in the following statement:
kafka-topics.sh –create –zookeeper localhost:2181 –replication-factor 3 –parti‐
tions 3 –topic bigdata_tweets.
Bigdata_tweets represents the flume sink, it consumes the event data and remove it
from the channel and act as storage for these messages that transit. Taking into account
the proprieties of different components cited above we deploy the flume agent using a
customized flume-agent configuration (see Fig. 4). The required jar files corresponding
to the source and sink are added to the library folder of flume in order to interact with
them.
Fig. 4. Sample of Flume agent configuration
Data Processing
This phase consists to ingest data from Kafka topic for live processing in Apache Spark.
Since Spark is a batch processing, we use Spark streaming to retrieve continuously the
messages accumulated in Kafka topic. Spark streaming receives the input stream and
divides it in a series of mini batches corresponding to input periods equal to batch
interval, it creates a DStream (see Fig. 5) which is as a sequence of RDDs that can be
processed in Spark core as a static data.
Fig. 5. Discretized data stream
Any streaming application needs a streaming context which is an entry point to the
Spark cluster resources.
We create our application in Scala that involves the following steps:
(1) To interact with kafka cluster, we connect spark streaming adopting the direct
approach using the DirectStream method in order to deploy a customized receiver
(see Fig. 6) which requires the subscription to bigdata_tweets topic created above.
Fig. 6. Spark streaming receiver
(2) Once the stream is created we convert it to JSON format (see Fig. 7), in order to
extract and process the interested fields in future analysis tasks. We store the stream
data in HDFS in JSON Format.
Fig. 7. Persisting the stream data in HDFS
Insights
Exploratory Analysis
We collected 20058 tweets, that we stored in HDFS in JSON format, then we converted
them to DataFrame in a structured format appropriate to be queried. We create a table
by selecting the entities and fields in interest like text, hash tags, urls, place, user.lang
in order to extract insights using Spark SQL. Thus, we deduced that the tweets contain
several links to a diversified resources for informal learning which can adapt to all styles
of learning in the form of links to external pages, free tutorial and courses (see Table 1).
We have noticed the presence of several companies specialized in the eLearning industry
which publish their offers and course promotions to attract users interested in big data
technologies and data science.
We found 9214 distinct users, although geo-location is disabled in the majority of
tweets [14], but we can extract their origin from the time zone, and native languages,
we found that 80% of users are Americans.
4264 distinct hashtags found in tweets data, we extract the top 10 most popular
hashtags (see Fig. 8) with respectively the number of occurrences in all tweets.
Table 1. Summary of links to external resources

Topics Total links to learning resources
Big data 157
Data science 84
Machine learning 408
Hadoop 70
Spark 390
Analytics 235
Fig. 8. Top 10 most popular hashtags.
Graph Data Structure and Finding Top Group of Connected Hashtags

Generally the raw data transformed for analysis tasks (see Fig. 9) are a set of records
stored in a table or a DataFrame, they are structured and divided in two dimensions
which are column and row.
Fig. 9. Sample of DataFrame created from raw data containing tweet identifier, user and hashtags.
In graph theory [19], graph is a data structure, conceptually described by a pair (S,
A) where S is a finite set of nodes called vertices or vertex and A is a finite multi-set of
ordered pairs of vertices called edges, an edge connects two vertices in a graph.
In real life applications, everything is interconnected, Graphs are mostly used to
represent the networks and model the relations between nodes, like routers, airports,
paths in cities, users in social networks.
A graph can be:
• Directed: the edges have a direction from the vertex source to the vertex destination
• Undirected: the edges have no direction.
• Directed multigraph: a pair of vertices is linked by one two or more edges, it describes
a multiple relationships. The edges share the same source and destination.
• Property Graphs: is a directed multigraph where vertex and edges have proprieties.
A tweet can contain 0 to multiple hashtags, each hashtag represents a topic of discus‐
sion, the presence of multiple hashtags increase the engagement of the users and the
value of the publication. Using Scala, we implement a graph analytics pipeline with
Spark Graph X in order to convert the DataFrame (as shown in Fig. 9) to a graph and
find the top connected hashtags.
Building a graph with Graph X requires two arguments: RDD of Vertices and RDD
of edges, which can be instantiated based on two specialized RDD implementations:
– The VertexRDD[VD] is a parameterized class, it’s defined as RDD[(VertexId, VD)],
VertexId is a vertex identifier, it is an instance of Long, VD is the vertex attribute or
property it can be a user type defined or other type of data information that are related
to vertex.
– The EdgeRDD[ED] is a parameterized class which is an implementation of
RDD[Edge[ED]], an instance of Edge represents VertexId source, VertexId destina‐
tion, and the attribute of the property of the edge.
We build the structure of vertices from the hashtag name, for each hashtag we create
a unique identifier (VertexId) in 64 bit by using the MurmurHash3 library [20], the vertex
propriety takes the string value of the hashtag name.
For the edge which is the link between two nodes, a pair of hashtags is generated by
using the combinations function, since we have no information about the relationship
between hashtags except their presence in the same tweet we opt to use the Twitter
username as propriety of the edge. A triplet represents an edge with two connected
vertices. We employed data with the hashtags entities having a size greater or equal to
2 to avoid the appearance of isolated nodes in our graph. We present as follow (see
Fig. 10) the steps to generate the structures of vertices and edges:
Fig. 10. Steps to generate the vertices and edges.
From a pair of RDD vertices and edges, we create an instance of Graph class to
generate a graph data structure as follows: val graph = Graph(vertices, edges) (Fig. 11).
Fig. 11. Sample of graph vertices, graph edges and graph triplets.
Total of vertices = 3329, Total of edges = 208973, Total of triplets = 208973

Connected component is a subgraph whose vertices is a subset of the set of vertices
of the original graph and whose edges is a subset of the set of the original graph. In
nutshell connected component is a subgraph whose vertices are interconnected by a set
of edges, if a vertex A is not linked directlty or indirecty to vertex B via another vertex
C, then A and B aren’t in the same connected component (Fig. 12).
Fig. 12. Sample of total vertices per component.
Connected components are generated by using the connectedComponents method

as follows: val connectedComponentsGraph = graph.connectedComponents
We extract the total of vertices and respectively the components to which they belong
as follows: connectedComponentsGraph.vertices.map(_._2).countByValue.toSeq.
sortBy(_._2). reverse.take(10)foreach(println)
A top group of connected component is performed using an InnerJoin method in
order to join vertices of the original graph and the vertices of the connected Components
based on VertexId, then we can filter the hashtags that belong to the component number
1, the result can be stored as a text file (see Fig. 13). The top group of connected compo‐
nent contains 3078 hashtags, which represents 92.40% of all the original graph vertices,
they are strong interrelated to our six topics: big data, data science, machine learning,
hadoop, spark and analytics.
Fig. 13. Sample of hashtags that belong to the top group of connected component
5 Conclusion
In this paper we propose a social network analysis system designed around a Twitter
API source. This system is in the form of a real time data pipeline capable to capture
events which are the tweets related to informal learning and categorize them in topics
in order to extract valuable information. We combine Apache Flume and Kafka to build
the data ingestion layer which is responsible to retrieve live data. Apache Kafka cluster
is used for categorizing the data that transit. To process data in real time we use Spark
Streaming library. HDFS is used as a persistence layer.
This work is based on a real experience where we have collected a dataset of 20058
tweets, then we accomplished some steps to achieve the data pipeline analysis, and
finally we extracted the top group of connected hashtags using Spark Graph X API.
During this work we have identified new directions concerning the eLearning. The
first is to study the use of social network platforms by Moroccan students for informal
learning purposes, and the second is to study how to integrate social networks channels
in formal learning settings like eLearning platforms.
References
1. Cameron, R., Harrison, J.L.: The interrelatedness of formal, non-formal and informal
learning: evidence from labour market program participants. Aust. J. Adult Learn. 52(2), 277–
309 (2012)
2. McPherson, M., Budge, K., Lemon, N.: New practices in doing academic development:
Twitter as an informal learning space. Int. J. Acad. Dev. 20(2), 126–136 (2015)
3. Wadhwa, P., Bhatia, M.P.S.: Social networks analysis: trends, techniques and future
prospects. In: Fourth International Conference on Advances in Recent Technologies in
Communication and Computing (ARTCom 2012), Bangalore, India, pp. 1–6 (2012)
4. White, T.: Hadoop: The Definitive Guide. O’Reilly Media, Inc., Newton (2012)
5. Zaharia, M., Chowdhury, M., Franklin, M.J., Shenker, S., Stoica, I.: Spark: cluster computing
with working sets. In: Proceedings of the 2nd USENIX Conference on Hot Topics in Cloud
Computing (2010)
6. Ghemawat, S., et al.: The Google File System. ACM SIGOPS Operating Systems Review
(2013)
7. Zaharia, M., et al.: Resilient distributed datasets: a fault-tolerant abstraction for in-memory
cluster computing. In: Proceedings of 9th USENIX Conference on Networked Systems
Design and Implementation, p. 2 (2012)
8. Dean, J., Ghemawat, S.: MapReduce: simplified data processing on large clusters. Commun.
ACM 51(1), 107–113 (2008)
9. Bonchi, F., Castillo, C., Gionis, A., Jaimes, A.: Social network analysis and mining for
business applications. ACM Trans. Intell. Syst. Technol. (TIST) Arch. 2(3), 37 (2011). Article
22
10. Kim, Y., Hwang, E., Rho, S.: Twitter news-in-education platform for social collaborative and
flipped learning. J. Supercomput. Springer, 1–19 (2016). https://doi.org/10.1007/
s11227-016-1776-x
11. Aramo-Immonen, H., Kärkkäinen, H., Jussila, J.J., Joel-Edgar, S., Huhtamäki, J.: Visualizing
informal learning behavior from conference participants’ Twitter data with the Ostinato
model. J. Comput. Hum. Behav. Arch. 55(PA), 584–595 (2016)
12. Huhtamäki, J., Russell, M.G., Rubens, N., Still, K.: Ostinato: the exploration-automation cycle of
user-centric, process-automated data-driven visual network analytics. In: Matei, S., Russell, M.,
Bertino, E. (eds.) Transparency in Social Media, pp. 197–222. Cham, Computational Social
Sciences, Springer (2015). https://doi.org/10.1007/978-3-319-18552-1_11
13. Chen, X., Vorvoreanu, M., Madhavan, K.: Mining social media data for understanding
students’ learning experiences. IEEE Trans. Learn. Technol. 7(3), 246–259 (2014)
14. Chandra, S., Khan, L., Muhaya, F.B.: Estimating Twitter user location using social
interactions–a content based approach. In: IEEE Third International Conference on Privacy,
Security, Risk and Trust and 2011 IEEE Third International Conference on Social Computing,
Boston, MA, pp. 838–843 (2011)
15. Twitter libraries homepage. https://developer.twitter.com/en/docs/developer-utilities/
twitter-libraries. Accessed 24 Feb 2018
16. Shreedharan, H.: Using Flume. O’Reilly Media, Inc., Sebastopol (2014)
17. Vohra, D.: Apache kafka. In: Practical Hadoop Ecosystem. Apress, Berkeley, CA Apache
(2016)
18. Apache Flume homepage. https://flume.apache.org/. Accessed 24 Feb 2018
19. Bondy, J.A., Murty, U.S.R.: Graph Theory with Applications. American Elsevier Publishing
Company, New York (1976)
20. MurmurHash3 documentation. https://www.scala-lang.org/files/archive/api/2.11.0-M4/
index.html#scala.util.hashing.MurmurHash3$. Accessed 24 Feb 2018
A MapReduce-Based Adjoint Method to
Predict the Levenson Self Report
Psychopathy Scale Value
Manal Zettam(B) , Jalal Laassiri, and Nourdddine Enneya
Informatics, Systems and Optimization Laboratory, Department of Computer

Science, Faculty of Science, Ibn Tofail University, Kenitra, Morocco
{manal.zettam,laassiri,enneya}@uit.ac.ma
Abstract. The Levenson Self Report Psychopathy serves as a measure

to spot persons with psychopathic disorders able to commit crime or
offend others. Indeed, predicting the Levenson Self Report Psychopathy
factors would help investigator and even psychologist to spot offenders.
In this paper, a statistical model is performed with the aim of predicting
the Levenson Self Report Psychopathy scale value. For this purpose, the
multiple regression statistical method is used. In addition, a parallelized
algebraic adjoint method is performed to solve the least square prob-
lem. The MapReduce framework is used for this purpose. The Apache
implementation of Mapreduce developed in Java untilled Hadoop 2.6.0
is deployed to tackle experiments.
Keywords: Levenson Self Report Psychopathy scale · MapReduce

HDFS · Multiple regression analysis · Prediction
1 Introduction
Psychopathy refers to a disorder characterized by antisocial behaviors and

exploitative interpersonal relationships [1,19]. According to [2], psychopathic
traits involve manipulative and callous use of others, shallow and short-lived
affect, irresponsible and impulsive behavior, egocentricy and pathological lying.
Nonetheless, psychopaths lack of basic prosocial personality traits such as empa-
thy, guilt, and perspective-taking [3–6]. Psychopaths generally exhibit glibness,
superficial charm, grandiosity and deception [4,19].
In literature several measures have been developed to assess psychopathic
personality traits [1]. The Hare psychopathic Checklist-Revised (PCL-R) and
The Levenson Self Report Psychopathy (LSRP) are the most widely used mea-
sures to assess psychopathic personality traits. The PCL-R measure was devel-
oped on a criminal population and showed a strong reliance on corroborating file
data thereby PCL-R measure is not appropriate for use in non-incarcerated sam-
ples. In contrast with PCL-R, the LSRP measure was developed on a collegial
population it is appropriate for use in non-incarcerated samples.
c Springer Nature Switzerland AG 2018
Y. Tabii et al. (Eds.): BDCA 2018, CCIS 872, pp. 16–28, 2018.
https://doi.org/10.1007/978-3-319-96292-4_2
Another random document with
no related content on Scribd:
slight but extremely vivid side-lights, but which (having been written
on shipboard) may perhaps be taken with the grain of palliative salt
which should frequently be cast upon the condemnatory utterances
of sea-weary, if not sea-sick, passengers on the raging deep when
they regard everything connected with the odious ship which
confines them. We are introduced to this colonial woman of affairs in
the sub-title of the journal, which states that the journey to New
Netherland was made “in a small Flute-ship called the Charles, of
which Thomas Singleton was Master; but the superior Authority over
both Ship and Cargo was in Margaret Filipse, who was the Owner of
both, and with whom we agreed for our Passage from Amsterdam to
New York, in New Netherland, at seventy-five Guilders for each
Person, payable in Holland.”
This “Margaret Filipse” was the daughter of Adolph Hardenbrook
who settled in Bergen, opposite New Amsterdam. She was the
widow of the merchant trader Peter Rudolphus De Vries when she
married Frederick Philipse. Her second husband was a carpenter by
trade, who worked for Governor Stuyvesant; but on his marriage with
the wealthy Widow De Vries, he became her capable business
partner, and finally was counted the richest man in the colony. She
owned ships running to many ports, and went repeatedly to Holland
in her own ships as supercargo. She was visited by Dankers in
Amsterdam in June, 1679. According to the custom of his religious
sect, he always called her by her Christian name, and wrote of her
as Margaret. He says:—
“We spoke to Margaret, inquiring of her when the ship
would leave. She answered she had given orders to have
everything in readiness to sail to-day, but she herself was of
opinion it would not be before Monday. We offered her the
money to pay for our passage, but she refused to receive it at
that time, saying she was tired and could not be troubled with
it that day.”
They waited patiently on shipboard for several days for Madam
Philipse to embark, and at last he writes:—
“We were all very anxious for Margaret to arrive, so that we
might not miss a good wind. Jan and some of the other
passengers were much dissatisfied. Jan declared, ‘If this wind
blows over I will write her a letter that will make her ears
tingle.’”
Landing at an English port, the travellers bought wine and vinegar,
“for we began to see it would go slim with us on the voyage,” and
Margaret bought a ship which was made ready to go to the Isle of
May and then to the Barbadoes. Over the purchase and equipment
of this ship arose a great quarrel, for “those miserable, covetous
people Margaret and her husband” tried to take away the Charles’
long-boat because timber for a new one was cheaper in New York
than in Falmouth, England. Naturally, the passengers objected to
crossing the Atlantic without a ship’s-boat. Dankers complained
further of Margaret’s “miserable covetousness,”—that she made the
ship lay to for an hour and a half and sent out the jolly-boat to pick
up a ship’s mop or swab worth six cents; and the carpenter swore
because she had not furnished new leather and spouts for the
pumps. Dankers explained at length the enhancement of the Philipse
profits through some business arrangement and preferment with the
Governor, by which Frederick Philipse became the largest trader with
the Five Nations at Albany, had a profitable slave-trade with Africa,
and, it is asserted, was in close bonds with the Madagascar pirates.
Whether “Margaret” favored this trade with the pirates is not known;
but it could probably be said of her trade, as of many others in the
colony, that it was hard to draw the dividing line between privateering
and piracy.
Her calling was not singular in New Amsterdam. The little town
abounded in women-traders.
Elizabeth Van Es was the daughter of one of the early Albany
magistrates. She married Gerrit Bancker, and on becoming a widow
removed to New York, where she promptly opened a store on her
own account, and conducted it with success till her death, in 1694. In
the inventory of her effects were a share in a brigantine, a large
quantity of goods and peltries, as well as various silver-clasped
Bibles, gold and stone rings, and silver tankards and beakers,
showing her success in her business career. The wife of the great
Jacob Leisler, a Widow Vanderveen when he married her, was a
trader. Lysbet, the widow of Merchant Reinier, became the wife of
Domine Drisius, of New York. She carried on for many years a
thriving trade on what is now Pearl Street, near Whitehall Street, and
was known to every one as Mother Drisius. The wife of Domine Van
Varick also kept a small store, and thus helped out her husband’s
salary.
Heilke Pieterse was the wife of the foremost blacksmith of New
Amsterdam; and as he monopolized the whole business of Long
Island, he died very rich,—worth at least ten thousand dollars. Not
overwhelmed or puffed up with the inheritance of such opulence,
Heilke carried on her husband’s business for many years with
success.
Margaret Backer was another successful business woman. For
years she acted as attorney for her husband while he was in foreign
countries attending to that end of his great foreign trade. Rachel
Vinje, involved in heavy lawsuits over the settlement of an estate,
pleaded her own case in court, and was successful. Women were
constant in their appearance in court as parties in contracts and
agreements.
The Schuyler family did not lack examples of stirring women-kind.
Margaret van Schlictenhorst, wife of the first Peter Schuyler, being
left a widow, managed her husband’s estate in varied business lines
with such thrift and prudence that in her will, made at eighty years of
age, she could assert that the property had vastly increased. She
was not out of public affairs, for during the Leisler troubles she was
the second largest subscriber to the fund in support of the
government; and she also lent money to pay the borrowed soldiers.
Her niece, Heligonda van Schlictenhorst, a shrewd spinster, was a
merchant, and furnished public supplies. The daughter of Peter
Schuyler married John Collins. A letter of his, dated 1722, shows her
capacity. I quote a clause from it:—
“Since you left us my wife has been in the Indian country,
and Van Slyck had purchased what he could at the upper end
of the land; she purchased the rest from Ignosedah to his
purchase. She has gone through a great deal of hardship and
trouble about it, being from home almost ever since you left
us; and prevailed with the Indians whilst there with trouble
and expense to mark out the land where the mine is into the
woods. Mrs. Feathers has been slaving with her all this while,
and hard enough to do with that perverse generation, to bring
them to terms.”
The picture of these two women in the wilds, treating and
bargaining and trading with the savages, seems curious enough to
us to-day. Women seem to have excelled in learning the Indian
languages. The daughter of Anneke Jans was the best interpreter in
the colony, and served as interpreter to Stuyvesant during his
famous treaty with the Six Nations.
Many of the leading taverns or hostelries were kept by women,—a
natural calling, certainly, for good housewives. Madam Van Borsum
was mistress of the Ferry Tavern in Breucklen. Annetje Litschar kept
the tavern which stood near the present site of Hanover Square.
Metje Wessell’s hostelry stood on the north side of Pearl Street, near
Whitehall Street.
More successful still and bold in trade was Widow Maria Provoost.
Scarce a ship came into port from Holland, England, the
Mediterranean, West Indies, or the Spanish Main, but brought to her
large consignments of goods. Her Dutch business correspondence
was a large one. She, too, married a second time, and, as Madam
James Alexander, filled a most dignified position, and became the
mother of Lord Stirling.
In a letter written by her husband, James Alexander, to his brother
William, and dated October 21, 1721, there is found a passage
which gives extraordinary tribute to her business capacity and her
powers of endurance alike. It reads thus:—
“Two nights agoe at eleven o’clock, my wife was Brought to
bed of a Daughter and is in as good health as can be
Expected, and does more than can be Expected of any
woman, for till within a few hours of her being brought to bed
She was in her Shop, and ever Since has given the price of
Goods to her prentice, who comes to her and asks it when
Customers come in. The very next day after She was brought
to bed she Sold goods to above thirty pounds value. And here
the business matters of her Shop which is Generally
Esteemed the best in New York, she with a prentice of about
16 years of age perfectly well manages without the Least help
from me, you may guess a little of her success.”
He closes his letter with a eulogy which can be cordially endorsed
by every reader:
“I must say my fortune in America is above my Expectation,
and I think even my Deserts, and the greatest of my good
fortune is in getting so Good a Wife as I have, who alone
would make ae man easy and happy had he nothing else to
depend on.”
Madam Alexander accumulated great wealth, and spent it
handsomely. She was the only person in town, besides the
Governor, who kept a coach. Her will is an interesting document, and
shows a fine style of housekeeping. The enumeration of great and
lesser drawing-rooms, front and back parlors, blue and gold leather
room, green and gold leather room, tapestry room, chintz room, etc.,
show its pretension and extent. She lived on Broad Street, had a fine
garden laid out in the Dutch taste, a house full of servants, and spent
her money freely as she made it thriftily. A very good portrait of her
exists. It shows an interesting countenance, with fine features, a
keen eye, and indicating robust health. She is not dressed with great
elegance, wearing the costume of the day,—a commonplace frilled
cap, folded kerchief, close sleeves, such as we are familiar with in
portraits of English women of her time.
Jane Colden, the daughter of Governor Cadwallader Colden, was
of signal service, not in trade, but in science. A letter written by her
father explains her interest and usefulness:—
“Botany is an amusement which may be made agreeable to
the ladies who are often at a loss to fill up their time. Their
natural curiosity and the pleasure they take in the beauty and
variety of dress seem to fit them for it.
“I have a daughter who has an inclination to reading, and a
curiosity for Natural Philosophy or Natural History, and a
sufficient curiosity for attaining a competent knowledge. I took
the pains to explain Linnæus’ system, and to put it into an
English form for her use by freeing it from technical terms,
which was easily done, by using two or three words in the
place of one. She is now grown very fond of the study, and
has made such a progress in it as, I believe, would please
you, if you saw her performance. Though she could not have
been persuaded to learn the terms at first, she now
understands to some degree Linnæus’ characters,—
notwithstanding she does not understand Latin. She has
already a pretty large volume in writing of the description of
plants. She has shewn a method of taking the impression of
the leaves on paper with printers’ ink, by a simple kind of
rolling press which is of use in distinguishing the species. No
description in words alone, can give so clear an idea, as when
assisted with a picture. She has the impression of three
hundred plants in the manner you’ll see by the samples. That
you may have some conception of her performance, and her
manner of describing, I propose to enclose some samples in
her own writing, some of which I think are new genera.”
Peter Collinson said she was the first lady to study the Linnæan
system, and deserved to have her name celebrated; and John Ellis,
writing of her to Linnæus in 1758, asks that a genus be named, for
her, Coldenella. She was also a correspondent of Dr. Whyte of
Edinburgh, and many learned societies in Europe. Walter Rutherfurd
enumerates her talents, and caps them with a glowing tribute to her
cheese-making.
We find the women of the times full of interest in public affairs and
active in good works. In the later days of the province, we learn of
the gifts to the army at Crown Point in 1755. In those days the
generous farmers of Queens County, Long Island, collected one
thousand and fifteen sheep, and these were “cheerfully given.”
“While their husbands at Great Neck were employed in getting
sheep, the good mothers in that neighborhood in a few hours
collected nearly seventy good large cheeses, and sent them to New
York to be forwarded with the sheep to the army.” Kings County
defrayed the expense of conveying these sheep and cheeses to the
army; and a letter of gratitude was promptly returned by the
commander-in-chief. Sir William Johnson, who said,—
“This generous humanity is unanimously and gratefully
applauded here by all. We pray that your benevolence may
be returned to you by the great Shepherd of the human kind a
thousand fold. And may those amiable housewives to whose
skill we owe the refreshing cheeses long continue to shine in
their useful and endearing stations.”
Kings County and Suffolk also sent cheeses, and we learn also:—
“The Women of County Suffolk ever good in such
Occasions are knitting several large bags of stockings and
mittens to be sent to the poorer soldiers at Forts William
Henry and Edward.”
In studying the history of the province, I am impressed with the
debt New Yorkers of Dutch descent owe, not to their forefathers, but
to their foremothers; the conspicuous decorum of life of these
women and their great purity of morals were equalled by their good
sense and their wonderful capacity in both domestic and public
affairs. They were as good patriots as they were good business
women; and though they were none of them what Carlyle calls
“writing-women,” it was not from poverty of good sense or natural
intelligence, but simply from the imperfection of their education
through lack of good and plentiful schools, and also want of stimulus
owing to absence of literary atmosphere.
A very shrewd woman-observer, writing in the middle of the
eighteenth century of the Dutch, gives what seems to me a very just
estimate and good description of one of their traits. She says:
“Though they have no vivacity, they are smarter, a great deal
smarter, than the English, that is, more uptaking.” Those who know
the exact Scotch meaning of “uptaking,” which is somewhat
equivalent to Anthony Trollope’s “observation and reception,” will
understand the closeness of the application of the term to the Dutch.
The Dutch women especially were “uptaking;” adaptive of all
comfort-bringing methods of housekeeping. This was noted by
Guicciardini in Holland as early as 1563. They were far advanced in
knowledge and execution of healthful household conditions, through
their beautiful cleanliness. Irving says very truthfully of them: “In
those good days of simplicity and sunshine a passion for cleanliness
was the leading principle in domestic economy, and the universal
test of a good housewife.” Kalm says: “They are almost over nice
and cleanly in regard to the floor, which is frequently scoured twice a
week.” They found conditions of housekeeping entirely changed in
America, but the passionate love of cleanliness fostered in the
Fatherland clung long in their hearts. Their “Œconomy” and thrift
were also beautiful.
An advertisement in the “New York Gazette” of April 1, 1751,
shows that the thrift of the community lingered until Revolutionary
times:—
“Elizabeth Boyd gives notice that she will as usual graft
Pieces in knit Jackets and Breeches not to be discern’d, also
to graft and foot Stockings, and Gentlemens Gloves, mittens
or Muffatees made out of old Stockings, or runs them in the
Heels. She likewise makes Childrens Stockings out of Old
Ones.”
Other dames taught more elegant accomplishments:—
“Martha Gazley, now in the city of New York, Makes and
Teacheth the following curious Works, viz.: Artificial Fruit and
Flowers and other Wax-Work, Nun’s Work, Philligree and
Pencil-work upon Muslin, all sorts of Needlework, and Raising
of Paste, as also to Paint upon Glass, and Transparent for
Sconces with other Works. If any young Gentlewomen, or
others, are inclined to learn any or all of the above-mentioned
curious Works, they may be carefully taught and instructed in
the same by said Martha Gazley.”
Mrs. Van Cortlandt, in her delightful account of home-life in
Westchester County, says of the industrious Dutch women and their
accomplishments and occupations:—
“Knitting was an art much cultivated, the Dutch women
excelling in the variety and intricacy of the stitches. A knitting
sheath, which might be of silver or of a homely goose-quill,
was an indispensable utensil, and beside it hung the ball-pin-
cushion. Crewel-work and silk embroidery were fashionable,
and surprisingly pretty effects were produced. Every little
maiden had her sampler, which she began with the alphabet
and numerals following them with a Scriptural text or verse of
a metrical psalm. Then fancy was let loose on birds, beasts,
and trees. Most of the old families possessed framed pieces
of embroidery, the handiwork of female ancestors. Flounces
and trimmings for aprons worked with delicately tinted silks on
muslin were common. I have several yards of fine muslin
painted in the early days with full-blown thistles in the
appropriate colors. Fringe looms were in use, and cotton and
silk fringes were woven.”
Tape-looms were also found in many households; and the weaving
of tapes and “none-so-prettys” was deemed very light and elegant
work.
Though to the Dutch is ascribed the invention of the thimble, I
never think of the Dutch women as excelling in fine needlework; and
I note that the teachers of intricate and novel embroidery-stitches are
always Englishwomen; but in turn the English goodwives must yield
to the Dutch the palm of comfortable, attractive housewifery, as well
as shrewd, untiring business capacity.
CHAPTER IX
THE COLONIAL WARDROBE
The Dutch goodwife worked hard from early morn till sunset. She
worked in restricted ways, she had few recreations and pleasures
and altogether little variety in her life; but she possessed what
doubtless proved to her in that day, as it would to any woman in this
day, a source of just satisfaction, a soothing to the spirit, a staying of
melancholy, a moral support second only to the solace of religion,—
namely, a large quantity of very good clothes, which were
substantial, cheerful, and suitable, if not elegant.
The Dutch never dressed “in a plaine habbit according to the
maner of a poore wildernesse people,” as the Connecticut colonists
wrote of themselves to Charles II.; nor were they weary wanderers in
a wilderness as were Connecticut folk.
I have not found among the statutes of New Netherland any
sumptuary laws such as were passed in Connecticut,
Massachusetts, and Virginia, to restrain and attempt to prohibit
luxury and extravagance in dress. Nor have I discovered in the court-
records any evidences of magisterial reproof of finery; there is, on
the contrary, much indirect proof of encouragement to “dress orderly
and well according to the fashion and the time.” Of course the Dutch
had no Puritanical dread of over-rich garments; and we must also
never forget New Netherland was not under the control of a
government nor of a religious band, but of a trading-company.
The ordinary dress of the fair dames and damsels of New
Amsterdam has been vividly described by Diedrich Knickerbocker;
and even with the additional light upon their wardrobe thrown by the
lists contained in colonial inventories, I still think his description of
their every-day dress exceedingly good for one given by a man. He
writes:
“Their hair, untortured by the abominations of art, was
scrupulously pomatumed back from their foreheads with a
candle, and covered with a little cap of quilted calico, which
fitted exactly to their heads. Their petticoats of linsey-woolsey
were striped with a variety of gorgeous dyes, though I must
confess those gallant garments were rather short, scarce
reaching below the knee; but then they made up in the
number, which generally equalled that of the gentlemen’s
small-clothes; and what is still more praiseworthy, they were
all of their own manufacture,—of which circumstance, as may
well be supposed, they were not a little vain.
“Those were the honest days, in which every woman
stayed at home, read the Bible, and wore pockets,—ay, and
that, too, of a goodly size, fashioned with patchwork into
many curious devices, and ostentatiously worn on the
outside. These, in fact, were convenient receptacles where all
good housewives carefully stored away such things as they
wished to have at hand; by which means they often came to
be incredibly crammed.
“Besides these notable pockets, they likewise wore scissors
and pincushions suspended from their girdles by red ribbons,
or, among the more opulent and showy classes, by a brass
and even silver chains, indubitable tokens of thrifty
housewives and industrious spinsters. I cannot say much in
vindication of the shortness of the petticoats; it doubtless was
introduced for the purpose of giving the stockings a chance to
be seen, which were generally of blue worsted, with
magnificent red clocks; or perhaps to display a well-turned
ankle and a neat though serviceable foot, set off by a high-
heeled leathern shoe, with a large and splendid silver buckle.
“There was a secret charm in those petticoats, which no
doubt entered into the consideration of the prudent gallants.
The wardrobe of a lady was in those days her only fortune;
and she who had a good stock of petticoats and stockings
was as absolutely an heiress as is a Kamtschatka damsel
with a store of bear-skins, or a Lapland belle with plenty of
reindeer.”
A Boston lady, Madam Knights, visiting New York in 1704, wrote:
—
“The English go very fashionable in their dress. But the
Dutch, especially the middling sort, differ from our women, in
their habitt go loose, wear French muches wch are like a
Capp and head-band in one, leaving their ears bare, which
are sett out with jewells of a large size and many in number;
and their fingers hoop’t with rings, some with large stones in
them of many Coullers, as were their pendants in their ears,
which you should see very old women wear as well as
Young.”
This really gives a very good picture of the vrouws; “loose in their
habit,” wearing sacques and loose gowns, not laced in with pointed
waists as were the English and Boston women; with the ornamental
head-dress, and the gay display of stoned earrings and rings, which
was also not the usual wear of New England women, who generally
owned only a few funeral rings.
In the inventories of personal estates contained in the Surrogate’s
Court we find details of the wardrobe; but as I have enumerated and
defined all the different articles at some length in my book, “Costume
of Colonial Times,” I will not repeat the definitions here; but it should
be remembered that in the enumeration of the articles of clothing,
many stuffs and materials of simple names were often of
exceedingly good and even rich quality. From those inventories we
have proof that all Dutch women had plenty of clothes; while the
wives of the burgomasters, the opulent merchants, and those in
authority, had rich clothes. I have given in full in my book a list of the
clothing of a wealthy New York dame, Madam De Lange; but I wish
to refer to it again as an example of a really beautiful wardrobe. In it
were twelve petticoats of varying elegance, some worth two pounds
fifteen shillings each, which would be more than fifty dollars to-day.
They were of silk lined with silk, striped stuff, scarlet cloth, and ash-
gray cloth. Some were trimmed with gold lace. With those petticoats
were worn samares and samares-a-potoso, six in number, which
were evidently jackets or fancy bodies; these were of calico, crape,
“tartanel,” and silk. One trimmed with lace was worth three pounds.
Waistcoats and bodies also appear; also fancy sleeves. Love-hoods
of silk and cornet-caps with lace make a pretty head-gear to
complete this costume, with which was worn the reim or silver girdle
with hanging purse, and also with a handsome number of diamond,
amber, and white coral jewels.
The colors in the Dutch gowns were almost uniformly gay,—in
keen contrast to the sad-colored garments of New England. Madam
Cornelia de Vos in a green cloth petticoat, a red and blue
“Haarlamer” waistcoat, a pair of red and yellow sleeves, and a purple
“Pooyse” apron was a blooming flower-bed of color.
The dress of Vrouentje Ides Stoffelsen, a very capable
Dutchwoman who went to Bergen Point to live, varied a little from
that of these town dames. Petticoats she had, and waistcoats,
bodies and sleeves; but there was also homelier attire,—purple and
blue aprons, four pairs of pattens, a fur cap instead of love-hoods,
and twenty-three caps. She wore the simpler and more universal
head-gear,—a close linen or calico cap.
The head covering was of considerable importance in New
Amsterdam, as it was in Holland as well as in England at that date.
We find that it was also costly. In 1665 Mistress Piertje Jans sold a
fine “little ornamental headdress” for fifty-five guilders to the young
daughter of Evert Duyckinck. But it seems that Missy bought this
“genteel head-clothes” without the knowledge or permission of her
parents, and on its arrival at the Duyckinck home Vrouw Duyckinck
promptly sent back the emblem of extravagance and disobedience.
Summoned to court by the incensed milliner who wished no rejected
head-dresses on her hands, and who claimed that the transaction
was from the beginning with full cognizance of the parents, Father
Duyckinck pronounced the milliner’s bill extortionate; and
furthermore said gloomily, with a familiar nineteenth-century
phraseology of New York fathers, that “this was no time to be buying
and wearing costly head-dresses.” But the court decided in the
milliner’s favor.
It is to be deplored that we have no fashion-plates of past
centuries to show to us in exact presentment the varying modes
worn by New York dames from year to year; that method of fashion-
conveying has been adopted but a century. The modes in olden days
travelled from country to country, from town to town, in the form of
dolls or “babies,” as they were called, wearing miniature model
costumes. These dolls were dressed by cutters and tailors in Paris or
London, and with various tiny modish garments were sent out on
their important mission across the water. In Venice a doll attired in
the last fashions—the toilette of the year—was for centuries
exhibited on each Ascension Day at the “Merceria” for the edification
of noble Venetian dames, who eagerly flocked to the attractive sight.
Not less eagerly did American dames flock to provincial mantua-
makers and milliners to see the London-dressed babies with their
miniature garments. Even in this century, fashions were brought to
New York and Philadelphia and Albany through “milliners’ boxes”
containing dressed dolls. Mrs. Vanderbilt tells of one much admired
fashion-doll of her youth who had a treasured old age as a juvenile
goddess.
A leading man of New Amsterdam, a burgomaster, had at the time
of his death, near the end of Dutch rule, this plentiful number of
substantial garments: a cloth coat with silver buttons, a stuff coat,
cloth breeches, a cloth coat with gimp buttons, a black cloth coat, a
silk coat, breeches and doublet, a silver cloth breeches and doublet,
a velvet waistcoat with silver lace, a buff coat with silk sleeves, three
“gross-green” cloaks, several old suits of clothes, linen, hosiery,
silver-buckled shoes, an ivory-headed cane, and a hat. One hat may
seem very little with so many other garments; but the real beaver
hats of those days were so substantial, so well-made, so truly worthy
an article of attire, that they could be constantly worn and yet last for
years. They were costly; some were worth several pounds apiece.
Gayer masculine garments are told of in other inventories: green
silk breeches flowered with silver and gold, silver gauze breeches,
yellow fringed gloves, lacquered hats, laced shirts and neck-cloths,
and (towards the end of the century, and nearly through the
eighteenth century) a vast variety of wigs. For over a hundred years
these unnatural abominations, which bore no pretence of resembling
the human hair, often in grotesque, clumsy, cumbersome shapes,
bearing equally fantastic names, and made of various indifferent and
coarse materials, loaded the heads and lightened the pockets of our
ancestors. I am glad to note that they were taxed by the government
of the province of New York. The barber and wig-maker soon
became a very important personage in a community so given over to
costly modes of dressing the head. Advertisements in the
newspapers show the various kinds of wigs worn in the middle of the
eighteenth century. From the “New York Gazette” of May 9, 1737, we
learn of a thief’s stealing “one gray Hair Wig, one Horse hair wig not
the worse for wearing, one Pale Hair Wig, not worn five times,
marked V. S. E., one brown Natural wig, One old wig of goat’s hair
put in buckle.” Buckle meant to curl; and derivatively a wig was in
buckle when it was rolled on papers for curling. Other
advertisements tell of “Perukes, Tets, and Fox-tails after the
Genteelest Fashion. Ladies’ Tets and wigs in perfect imitation of their
own hair.” Other curious notices are of “Orange Butter” for
“Gentlewomen to comb up their hair with.”
This use of orange butter as a pomatum was certainly unique; it
was really a Dutch marmalade. I read in my “Closet of Rarities,”
dated 1706:—
“The Dutch Way to make Orange-butter. Take new cream
two gallons, beat it up to a thicknesse, then add half a pint of
orange-flower-water, and as much red wine, and so being
become the thicknesse of butter it has both the colour and
smell of an orange.”
A very characteristic and eye-catching advertisement was this
from the “New York Gazette” of May 21, 1750:—
“This is to acquaint the Public, that there is lately arrived
from London the Wonder of the World, an Honest Barber and
Peruke Maker, who might have worked for the King, if his
Majesty would have employed him: It was not for the want of
Money he came here, for he had enough of that at Home, nor
for the want of Business, that he advertises hinself, BUT to
acquaint the Gentlemen and Ladies, That Such a Person is
now in Town, living near Rosemary Lane where Gentlemen
and Ladies may be supplied with Goods as follows, viz.: Tyes,
Full-Bottoms, Majors, Spencers, Fox-Tails, Ramalies, Tacks,
cut and bob Perukes: Also Ladies Tatematongues and Towers
after the Manner that is now wore at Court. By their Humble
and Obedient Servant,
“John Still.”
With the change from simple Dutch ways of hairdressing came in
other details more constrained modes of dressing. With the wig-
maker came the stay-maker, whose curious advertisements may be
read in scores in the provincial newspapers; and his arbitrary
fashions bring us to modern times.
From the deacons’ records of the Dutch Reformed Church at
Albany we catch occasional hints of the dress of the children of the
Dutch colonists. There was no poor-house, and few poor; but since
the church occasionally helped worthy folk who were not rich, we
find the deacons in 1665 and 1666 paying for blue linen for
schorteldoecykers, or aprons, for Albany kindeken; also for haaken
en oogen, or hooks and eyes, for warm under-waists called
borsrockyen. They bought linen for luyers, which were neither
pinning-blankets nor diapers, but a sort of swaddling clothes, which
evidently were worn then by Dutch babies. Voor-schooten, which
were white bibs; neerstucken, which were tuckers, also were worn
by little children. Some little Hans of Pieter had given to him by the
deacons a fine little scarlet aperock, or monkey-jacket; and other
children were furnished linen cosynties, or night-caps with capes.
Yellow stockings were sold at the same time for children, and a gay
little yellow turkey-legged Dutchman in a scarlet monkey-jacket and
fat little breeches must have been a jolly sight.
CHAPTER X
HOLIDAYS
The most important holidays of the early years of the colony were,
apparently, New Year’s Day and May Day, for we find them named
through frequent legislation about rioting on these days, repairing of
damages, etc. It has been said that New Yorkers owe to the Dutch
an everlasting gratitude for our high-stoop houses and the delights of
over two centuries of New Year’s calling. The latter custom lived long
and happily in our midst, died a lingering and lamented death, is still
much honored in our memory, and its extinction deeply deplored and
unwillingly accepted.
The observance of New Year’s Day was, without doubt, followed
by both Dutch and English from the earliest settlement. We know
that Governor Stuyvesant received New Year’s calls, and we also
know that he prohibited excessive “drunken drinking,” unnecessary
firing of guns, and all disorderly behavior on that day. The reign of
the English did not abolish New Year’s visits; and we find Charles
Wolley, an English chaplain, writing in his journal in New York in
1701, of the addition of the English custom of exchange of gifts:—
“The English in New York observed one anniversary custom
and that without superstition, I mean the strenarum
commercium, as Suetonius calls them, a neighborly
commerce of presents every New Year’s Day. Some would
send me a sugar-loaf, some a pair of gloves, some a bottle or
two of wine.”
A further celebration of the day by men in New York was by going
in parties to Beekman’s Swamp to shoot at turkeys.
New Year’s calling was a new fashion to General Washington
when he came to New York to live for a short time, but he adopted it
with approval; and his New Year’s Receptions were imposing
functions.
For a long time the New Year was ushered in, in country towns,
with great noise as well as rejoicing. All through the day groups of
men would go from house to house firing salutes, and gathering
gradually into large parties by recruits from each house until the end
of the day was spent in firing at a mark. The Legislature in March,
1773, attempted to stop the gun-firing, asserting that “great damages
are frequently done on the eve of the last day of December and on
the first and second days of January by persons going from house to
house with guns and other firearms.” In 1785 a similar enactment
was passed by the State Legislature.
In the palmiest days of New Year’s calling, New York City
appeared one great family reunion. Every wheeled vehicle in the
town seemed to be loaded with visitors going from house to house.
Great four and six horse stages packed with hilarious mobs of men
went to the house of every acquaintance of every one in the stage.
Target companies had processions; political bodies called on families
whose head was well known in political life. The newspaper-carriers
brought out addresses yards long with rhymes:—
“The day devoted is to mirth,

And now around the social hearth
Friendship unlocks her genial springs,
And Harmony her lyre now strings.
While plenty spreads her copious hoard,
And piles and crowns the festive board,”
etc., etc., for hundreds of lines.

The “copious hoard” of substantial food, with decanters of wine,
bowls of milk punch, and pitchers of egg-nog, no longer “crown the
festive board” on New Year’s Day; but we still have New Year’s
Cakes, though not delivered by singing bakers’ ’prentices as of yore.
May Day was observed in similar fashion,—by firing of guns, gay
visiting, and also by the rearing of maypoles.
A very early mention of a maypole is in June, 1645, when one
William Garritse had “sung a libellous song” against Rev. Francis
Doughty, the preacher at Flushing, Long Island, and was sentenced
in punishment therefor to be tied to the maypole, which in June was
still standing. Stuyvesant again forbade “drunken drinking,” and firing
of guns and planting of maypoles, as productive of bad practices. I
don’t know whether the delight of my childhood, and of generations
of children in Old and New England up to this present May Day on
which I am now writing,—the hanging of May baskets,—ever made
happy children in New York.
There was some observance in New York of Shrovetide as a
holiday-time. As early as 1657 we find the sober Beverwyck
burghers deliberating on “some improprieties committed at the house
of Albert de Timmerman on Shrovetide last.” As was the inevitable
custom followed by the extremely uninventive brain of the
seventeenth and eighteenth century rioter, were he Dutch or English,
these “improprieties” took the form of the men’s parading in women’s
clothes; Pieter Semiensen was one of the masqueraders. Two years
later the magistrates were again investigating the “unseemly and
scandalous” celebration of Shrovetide; and as ever before, the youth
of early Albany donned women’s clothes and “marched as
mountebanks,” as the record says, just as they did in Philadelphia
and Baltimore and even in sober Boston. We find also for sale in
Beverwyck at this time, noisy Shrovetide toys—rommelerytiens, little
“rumbling-pots,” which the youth and children doubtless keenly
enjoyed.
At an early date Shrovetide observances, such as “pulling the
goose,” were prohibited by Governor Stuyvesant in New York. A mild
protest on the part of some of the burgomasters against this order of
the Governor brought forth one of Stuyvesant’s characteristically
choleric edicts in answer, in which he speaks of having “interdicted
and forbidden certain farmers’ servants to ride the goose at the feast
of Backus and Shrovetide ... because it is altogether unprofitable,
unnecessary, and criminal for subjects and neighbors to celebrate
such pagan and popish feasts and to practise such customs,
notwithstanding the same may in some places of Fatherland be
tolerated and looked at through the fingers.” Domine Blom, of
Kingston or Wyltwyck, joined in the governor’s dislike of the game.
But there were some of the magistrates who liked very well to “pull
the goose” themselves, so it is said. It was a cruel amusement. The
thoroughly greased goose was hung between two poles, and the
effort of the sport was to catch, snatch away, and hold fast the poor
creature while passing at great speed. In Albany in 1677 all
“Shrovetide misdemeanors were prohibited, viz.: riding at a goose,
cat, hare, and ale.” The fine was twenty-five guilders in sea-want.
What the cat, hare, and ale part of the sport was, I do not know.
In New York by the middle of the eighteenth century Shrove
Tuesday was firmly assigned to cocking-mains. The De Lanceys
were patrons of this choice old English sport. Cock-gaffs of silver
and steel were freely offered for sale in New York and Maryland
newspapers, and on Shrove Tuesday in 1770 Jacob Hiltzeheimer
attended a famous cock-fight on the Germantown road. We cannot
blame honest New Yorkers if they did not rise above such rude
sports, when cock-fighting and cock-throwing and cock-squoiling and
cock-steling obtained everywhere in Old England at Shrovetide;
when school-boys had cock-fights in their school-rooms; and in
earlier days good and learned old Roger Ascham ruined himself by
betting on cock-fights, and Sir Thomas More boasted proudly of his
skill in “casting a cock-stele.”
Mr. Gabriel Furman, writing in 1846, told of an extraordinary
observance of Saint Valentine’s Day by the Dutch—one I think
unknown in folk-lore—which obtained on Long Island among the
early settlers. It was called Vrouwen dagh, or Women’s day, and was
thus celebrated: Every young girl sallied forth in the morning armed
with a heavy cord with knotted end. She gave to every young man
whom she met several smart lashes with the knotted cord. Perhaps
these were “love-taps,” and were given with no intent of stinging.
Judge Egbert Benson wrote, in 1816, that in New York this custom
dwindled to a similar Valentine observance by New York children,
when the girls chased the boys with many blows. In one school the

Ebffiledoc 3513

Uploaded by

Copyright:

Available Formats

Ebffiledoc 3513

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ebffiledoc 3513

Uploaded by

Copyright:

Available Formats

Big Data Cloud and Applications Third

International Conference BDCA 2018

Intelligence Science and Big Data Engineering 8th

Advances in Computing and Data Sciences Second

Advances in Computing and Data Sciences Second

Soft Computing Systems Second International Conference

Cloud Computing and Security 4th International

Cloud Computing and Security 4th International

Cloud Computing and Security 4th International

Cyber Physical Systems Model Based Design 8th

Communications in Computer and Information Science 872

Big Data, Cloud

Mohammed Al Achhab Nourddine Enneya (Eds.)

Big Data, Cloud

ISSN 1865-0929 ISSN 1865-0937 (electronic)

Library of Congress Control Number: 2018948223

© Springer Nature Switzerland AG 2018

July 2018 Youness Tabii

Local Organizing Committee

Kamal Baraka Cadi Ayyad University, Morocco

Ebroul Izquierdo Queen Mary, University of London, UK

Informal Learning in Twitter: Architecture of Data Analysis Workflow

A MapReduce-Based Adjoint Method to Predict the Levenson Self Report

Big Data Optimisation Among RDDs Persistence in Apache Spark . . . . . . . . 29

QoS in the Cloud Computing: A Load Balancing Approach

A Proposed Approach to Reduce the Vulnerability in a Cloud System . . . . . . 55

A Multi-factor Authentication Scheme to Strength Data-Storage Access . . . . . 67

A Novel Text Encryption Algorithm Based on the Two-Square

Improving Sentiment Analysis of Moroccan Tweets

Comparative Study of Feature Engineering Techniques

Business Process Instances Scheduling with Human Resources

Hashtag Recommendation Using Word Sequences’ Embeddings . . . . . . . . . . 131

Towards for Using Spectral Clustering in Graph Mining . . . . . . . . . . . . . . . 144

Automatic Classification of Air Pollution and Human Health . . . . . . . . . . . . 160

Deep Semi-supervised Learning for Virtual Screening Based

Using Deep Learning Word Embeddings for Citations Similarity

Using Unsupervised Machine Learning for Data Quality.

Advanced Machine Learning Models for Large Scale Gene Expression

Stemming and Lemmatization for Information Retrieval Systems

Mobility of Web of Things: A Distributed Semantic

Comparison of Feature Selection Methods for Sentiment Analysis. . . . . . . . . 261

A Hierarchical Nonlinear Discriminant Classifier Trained

A Feature Level Fusion Scheme for Robust Speaker Identification . . . . . . . . 289

One Class Genetic-Based Feature Selection for Classification

Multiobjective Local Search Based Hybrid Algorithm for Vehicle

Dimension Reduction Techniques for Signal Separation Algorithms . . . . . . . 326

A Probabilistic Vector Representation and Neural Network

Improving Implementation of Keystroke Dynamics Using K-NN

SARIMA Model of Bioelectic Potential Dataset . . . . . . . . . . . . . . . . . . . . . 367

New Starting Point of the Continuous Hopfield Network . . . . . . . . . . . . . . . 379

Information System And Social Media

A Concise Survey on Content Recommendations . . . . . . . . . . . . . . . . . . . . 393

Toward a Model of Agility and Business IT Alignment . . . . . . . . . . . . . . . . 406

Integration of Heterogeneous Classical Data Sources

Toward a Solution to Interoperability and Portability of Content

Image Processing and Applications

Reconstruction of the 3D Scenes from the Matching Between Image Pair

Similarity Performance of Keyframes Extraction on Bounded Content

Natural Language Processing