Distributed Multi Class SVM for Large Data Sets
Pages 54 - 58
Abstract
Data mining algorithms are originally designed by assuming the data is available at one centralized site. These algorithms also assume that the whole data is fit into main memory while running the algorithm. But in today's scenario the data has to be handled is distributed even geographically. Bringing the data into a centralized site is a bottleneck in terms of the bandwidth when compared with the size of the data. In this paper for multiclass SVM we propose an algorithm which builds a global SVM model by merging the local SVMs using a distributed approach(DSVM). And the global SVM will be communicated to each site and made it available for further classification. The experimental analysis has shown promising results with better accuracy when compared with both the centralized and ensemble method. The time complexity is also reduced drastically because of the parallel construction of local SVMs. The experiments are conducted by considering the data sets of size 100s to hundred of 100s which also addresses the issue of scalability.
References
[1]
C. R. Y. K. al. Prediction of conversion from mild cognitive impairment to alzheimer disease based on bayesian data mining with ensemble learning. The Neuroradiology Journal, 25(1), 2012.
[2]
G. Aruna, Ranjani, Aditi, and S. Sahay. A novel approach to distributed mutli-class svm. Transactions on Machine Learning and Artificial Intelligence, 2(5):72--79, October 2014.
[3]
J. Cervantes, X. Li, and W. Yu. Multi-class svm for large data sets considering models of classes distribution. In International Conference on Data Mining, pages 257--268. DMIN, July 2008.
[4]
H. Dutta and et al. Distributed top-k outlier detection from astronomy catalogs using the demac system. In SDM Proceedings, pages 473--476. SIAM, 2007.
[5]
A. Ghodselahi. A hybrid support vector machine ensemble model for credit scoring. International Journal of Computer Applications, 17(5):0975--8887, March 2011.
[6]
B. Han. X. Classification by pairwise coupling. Advances in Neural Information Processing, 10(2):291--301, June 1998.
[7]
B. Han. X. Dcmsvm: Distributed parallel training for single-machine multiclass classifiers. In Computer Vision and Pattern recognition Proceedings, pages 3554--3561. IEEE, June 2012.
[8]
D. K, K. Bhaduri, and et al. Scalable distributed change detection from astronomy data straems using local, asyncronous eigen monitoring algorithms. In SDM Proceedings, pages 247--258. SIAM, 2009.
[9]
H. Kargupta, C. Gianella, and K. Sivakumar. Distributed data mining for earth and space science applications. In Proceeding of the fourth annual Earth Science Technology conference (ESTC), June 2004.
[10]
H. C. Kob and G. Tan. Data mining applications in healthcare. Journal of Healthcare Information Management, 19(2):64--72.
[11]
R. Mallik, N. Sarda, and H. Kargupta. Distributed data mining for sustainable smart grids. In Proceedings of the Sustainable KDD Workshop. KDD 2011. ACM, 2011.
[12]
J. D. M. Rennie and R. Rifkin. Improving multiclass text classification with the support vector machine. Technical report, Massachusetts Institute of Technology, 2001.
[13]
C. S. Stefano Lodi, Ricardo Nanculef. Single-pass distributed learning of multi-class svms using core-sets. In SDM Proceedings, pages 257--268. SIAM, 2010.
[14]
P. -N. Tan, V. Kumar, and M. Steinbach. Introduction to Data Mining. Pearson, New Delhi, India, 2012.
[15]
J. Wang and H. Wang. Application of data mining in the financial data forecasting. Advanced Intelligent Computing Theories and Applications. With Aspects of Theoretical and Methodological Issues, Lecture Notes in Computer Science, 5226(1):954--961, 2008.
Index Terms
- Distributed Multi Class SVM for Large Data Sets
Recommendations
Multi-class Support Vector Machine (SVM) Classifiers -- An Application in Hypothyroid Detection and Classification
BIC-TA '11: Proceedings of the 2011 Sixth International Conference on Bio-Inspired Computing: Theories and ApplicationsThe paper presents a Multi-class Support Vector Machine classifier and its application to hypothyroid detection and classification. Support Vector Machines (SVM) have been well known method in the machine learning community for binary classification ...
A New Multi-class SVM Algorithm Based on One-Class SVM
ICCS '07: Proceedings of the 7th international conference on Computational Science, Part III: ICCS 2007Multi-class classification is an important and on-going research subject in machine learning and data mining. In this paper, we propose a new support vector algorithm, called OC-K-SVM, for multi-class classification based on one-class SVM. For k-class ...
A New Orientation for Multi-Class SVM
SNPD '07: Proceedings of the Eighth ACIS International Conference on Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing - Volume 03Combined by several binary-class SVMs, present Multi-Class SVMs are usually inefficient in training process. When there is large number of categories of data to classify, training it would be very difficult. Expanded from Hyper-Sphere One-Class SVM (...
Comments
Please enable JavaScript to view thecomments powered by Disqus.Information & Contributors
Information
Published In
August 2015
763 pages
ISBN:9781450333610
DOI:10.1145/2791405
- Editor:
- Indu Nair,
- General Chairs:
- Sushmita Mitra,
- Ljiljana Trajković,
- Program Chairs:
- Punam Bedi,
- Suzanne McIntosh,
- M. S. Rajasree
Copyright © 2015 ACM.
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 10 August 2015
Check for updates
Author Tags
Qualifiers
- Research-article
- Research
- Refereed limited
Conference
WCI '15
WCI '15: Third International Symposium on Women in Computing and Informatics
August 10 - 13, 2015
Kochi, India
Acceptance Rates
WCI '15 Paper Acceptance Rate 98 of 452 submissions, 22%;
Overall Acceptance Rate 98 of 452 submissions, 22%
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 102Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 21 Nov 2024
Other Metrics
Citations
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in