Papers by Ralph Sherwin Corpuz
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
Mindanao Journal of Science and Technology, 2021
Generating accurate and timely internal and external audit reports may seem difficult for some au... more Generating accurate and timely internal and external audit reports may seem difficult for some auditors due to limited time or expertise in matching the correct clauses of the standard with the textual statement of findings. To overcome this gap, this paper presents the design of text classification models using support vector machine (SVM) and long short-term memory (LSTM) neural network in order to automatically classify audit findings and standard requirements according to text patterns. Specifically, the study explored the optimization of datasets, holdout percentage and vocabulary of learned words called NumWords, then analyzed their capability to predict training accuracy and timeliness performance of the proposed text classification models. The study found that SVM (96.74%) and LSTM (97.54%) were at par with each other in terms of the best training accuracy, although SVM (67.96±17.93 seconds [s]) was found to be significantly faster than LSTM (136.67±96.42 s) in any dataset size. The study proposed optimization formulas that highlight dataset and holdout as predictors of accuracy, while dataset and NumWords as predictors of timeliness. In terms of actual implementation, both classification models were able to accurately classify 20 out of 20 sample audit findings at 1 and 3 s, respectively. Hence, the extent of choosing between the two algorithms depend on the datasets size, learned words, holdout percentage, and workstation speed. This paper is part of a series, which explores the use of Artificial Intelligence (AI) techniques in optimizing the performance of QMS in the context of a state university.
Bookmarks Related papers MentionsView impact
International Journal on Information Technologies & Security, 2021
Classifying unstructured text data written in natural languages is a cumbersome task, and this is... more Classifying unstructured text data written in natural languages is a cumbersome task, and this is even worse in cases of vast datasets with multiple languages. In this paper, the author explored the utilization of Long Short-Term Neural Network (LSTM) in designing a classification model that can learn text patterns and classify English and Tagalog-based complaints, feedbacks and commendations of customers in the context of a state university in the Philippines. Results shown that the LSTM has its best training accuracy of 91.67% and elapsed time of 34s when it is tuned with 50 word embedding size and 50 hidden units. The study found that the lesser the number of hidden units in the network correlates to a higher classification accuracy and faster training time, but word embedding size has no correlation to the classification performance. Furthermore, results of actual testing proven that the proposed text classification model was able to predict 19 out of 20 test data correctly, hence, 95% classification accuracy. This means that the method conducted was effective in realizing the primary outcome of the study. This paper is part of a series of studies that employs machine and deep learning techniques toward the improvement of data analytics in a Quality Management System (QMS).
Bookmarks Related papers MentionsView impact
International Journal of Integrated Engineering , 2021
Analyzing natural language-based Customer Satisfaction (CS) is a tedious process. This issue is p... more Analyzing natural language-based Customer Satisfaction (CS) is a tedious process. This issue is practically true if one is to manually categorize large datasets. Fortunately, the advent of supervised machine learning techniques has paved the way toward the design of efficient categorization systems used for CS. This paper presents the feasibility of designing a text categorization model using two popular and robust algorithms – the Support Vector Machine (SVM) and Long Short-Term Memory (LSTM) Neural Network, in order to automatically categorize complaints, suggestions, feedbacks, and commendations. The study found that, in terms of training accuracy, SVM has best rating of 98.63% while LSTM has best rating of 99.32%. Such results mean that both SVM and LSTM algorithms are at par with each other in terms of training accuracy, but SVM is significantly faster than LSTM by approximately 35.47s. The training performance results of both algorithms are attributed on the limitations of the dataset size, high-dimensionality of both English and Tagalog languages, and applicability of the feature engineering techniques used. Interestingly, based on the results of actual implementation, both algorithms are found to be 100% effective in accurately predicting the correct CS categories. Hence, the extent of preference between the two algorithms boils down on the available dataset and the skill in optimizing these algorithms through feature engineering techniques and in implementing them toward actual text categorization applications.
Bookmarks Related papers MentionsView impact
Makara Journal of Technology, 2020
Risk-based thinking (RBT) is one of the distinct new features of the International Organization f... more Risk-based thinking (RBT) is one of the distinct new features of the International Organization for Standardization 9001:2015. Interestingly, the standard does not prescribe any tools. Hence, organizations are puzzled as to the extent of conformance with the RBT requirements. Some organizations have adopted formal tools. However, these tools seem insufficient in linking the standard into an evidence-based decision support system. To resolve gaps in RBT implementation, this paper proposes a framework based on fuzzy inference system (FIS) and support vector machine (SVM) to automate risk analysis and evaluation, proposal and verification of action plans, and prediction of the feasibility of risks and opportunities according to text patterns. Modeling results indicate that the framework has no significant difference in terms of accuracy compared with the conventional method. Both FIS-1 and FIS-2 models, however, are statistically significantly faster at 3.26 and 1.15 s, respectively. Meanwhile, the SVM model, whose text classification features are not evident in the conventional method, has a 97.16% classification accuracy and 2.6% confusion error during training, and 95% classification accuracy during testing. Results affirm that FIS and SVM are efficient tools in feasibly conforming with the RBT requirements of the ISO 9001:2015 international standard. Abstrak Standardisasi (ISO) 9001: 2015. Menariknya, standar tidak menentukan alat apa pun, karenanya, organisasi bingung pada tingkat kesesuaian. Beberapa organisasi telah mengadopsi alat formal, namun, mereka tampaknya tidak cukup dalam menghubungkan standar ke dalam sistem pendukung keputusan berbasis bukti. Untuk mengatasi kesenjangan dalam implementasi RBT, makalah ini mengusulkan kerangka kerja yang didasarkan pada Fuzzy Inference System (FIS) dan Support Vector Machine (SVM) untuk mengotomatisasi analisis risiko dan evaluasi, proposal dan verifikasi rencana tindakan, dan untuk memprediksi kelayakan risiko. dan peluang sesuai dengan pola teks. Berdasarkan hasil pemodelan, kerangka kerja ditemukan tidak memiliki perbedaan yang signifikan dalam hal akurasi dibandingkan dengan metode konvensional. Kedua model FIS-1 dan FIS-2, secara statistik lebih cepat secara signifikan pada masing-masing 3,26 dan 1,15. Sementara itu, model SVM, yang fitur klasifikasi teks tidak jelas dalam metode konvensional, ditemukan memiliki akurasi klasifikasi 97,16% dan kesalahan kebingungan 2,6% selama pelatihan, dan akurasi klasifikasi 95% selama pengujian. Hasil ini menegaskan bahwa FIS dan SVM adalah alat yang efisien untuk secara layak sesuai dengan persyaratan RBT dari standar internasional ISO 9001: 2015.
Bookmarks Related papers MentionsView impact
The study focused on the development of a technology management framework intended for Live Chat ... more The study focused on the development of a technology management framework intended for Live Chat providers so they can help online business customers optimize their Live Chat utilization. The study involved two hundred twenty-nine Customer Service Representatives, Developers, Business Owners, and Web Masters who have utilized at least one of the most popular Live Chat providers like Zopim, Kayako, Live Person, Live Chat Inc, and Olark. Specifically, the Live Chat Optimization Model, the Software as a Service Model (SaaS) Live Chat Consultant Website, and the Critical Success Factors (CSF) with a tiered degree of solutions were proposed. The stakeholders evaluated the framework to be very relevant, effective, efficient, impactful, and sustainable. Overall they were very satisfied with the outcomes of the framework.
Bookmarks Related papers MentionsView impact
This research aimed to develop an online module that would revolutionize the learning motivation ... more This research aimed to develop an online module that would revolutionize the learning motivation of students in taking highly technical subjects such as Microchip in C Language (Microcontroller Applications Laboratory) based on the Analyze-Design-Develop-Engage-Evaluate (ADDEE) framework for online module development. The research was conducted based on descriptive and developmental research designs, conforming to the Content-Interactivity-Support (CIS) design model and then implemented using the Waterfall Methodology. The module was developed in its initial phase in order to device feasible set of solutions to the identified students, teacher, technology, and institutional support challenges through the integration of various content, data, records, classroom, assessment, communication, and support management systems in an independent or blended learning approach. The module was tested to be intuitive, mobile, real time, and interactive. The respondents, composed of 28 randomly selected students of the College of Industrial Technology, Technological University Manila Campus, and 12 instructional developers, evaluated the system and found out that it was functionally suitable (x=3.57), reliable (x=3.09), operable (x=3.91), performance-wise efficient (x =4.11), secured (x=2.99), compatible (x=3.55), maintainable (x=3.41), and transferable (x=3.99) based on the ISO 25010:2011 Evaluation System, hence, it significantly contributed to the level of achievement of the students based on their semifinal and final exam scores [t 1 (13)=-962, p 1 =0.011; t 2 (13)=-.3129; p 2 =0.008]. The study further recommended its continuation to Phase 2 through the utilization of dedicated online database, high-level security layer
Bookmarks Related papers MentionsView impact
This research aimed to design and develop an autonomous robot to feasibly address waste disposal ... more This research aimed to design and develop an autonomous robot to feasibly address waste disposal issues in common indoor places. The researchers explored opportunities to improve path planning using Fuzzy Logic Control (FLC). The researchers utilized a Microcontroller Unit (MCU) to control input proximity, sound, and infrared sensors, and output geared Direct Current (DC) motors through machine learning and electromechanical interface. The researchers simulated an adaptive algorithm using Mamdani-type FLC model, implemented using C programming language, then downloaded as machine code to a real prototype. Based on significant test results, the waste robot accurately detected human interference, a feature that would be pivotal in overcoming individual indifferences on waste management.
Bookmarks Related papers MentionsView impact
Fires are among the worst disasters that can happen anywhere, anytime, and to anyone, for so many... more Fires are among the worst disasters that can happen anywhere, anytime, and to anyone, for so many reasons. The risks are even higher for highly-congested areas and public buildings where evacuation schemes are not effectively planned and implemented. There is a wide-range of fire-sensing technologies available in the market to prevent fire incidents, however, the extent of flexibility and reliability of these devices are still subject for further improvements. This paper presents the design and development of a Fire Evacuation System (FES) using Mamdani-type Fuzzy Logic Control (FLC) with the main intent to improve responsiveness and reliability of fire detection. The authors compared two types of smoke sensors then interfaced with a microcontroller unit, relay drivers, Light Emitting Diodes (LED) indicators, and other system peripherals to simulate the existence of smoke and to indicate the evacuation map. The authors further implemented the FLC rules using C programming language in an Arduino Integrated Development Environment (IDE). Testing results revealed that there is no significant difference on the responsiveness of the two smoke sensors with independent T-test value of t (26.131) =-0.026 and p = 0.979. Likewise, test results further proved that there is no significant difference on the reliability performance of the smoke sensors and LED indicators with dependent T-test value of tA_ON (19) =-0.847; pA_ON = 0.408; tA_OFF (19) = 0.678; pA_OFF = 0.506; tB_ON (19) =-0.764; pB_ON = 0.454; tB_OFF (19) =1.212; pB_OFF = 0.240 in activated and deactivated modes, respectively. These results confirmed that the prototype is responsive and reliable, and the use of FLC is effective for the design and development of fire evacuation systems.
Bookmarks Related papers MentionsView impact
Auditors of Quality Management System (QMS) face challenges in generating accurate audit reports ... more Auditors of Quality Management System (QMS) face challenges in generating accurate audit reports due to some factors that can be attributed to technical competence, experience, time, auditee reaction, and other factors. Incorrect clauses cited in audit reports may result to loss of integrity of the auditor and the auditing procedure itself, hence, it is important that auditors should be careful in citing clauses of the standard to avoid chaos and complaints from auditees. To resolve this issue, this paper presents the implementation of Artificial Neural Network (ANN) using Scaled Conjugate Gradient (SCG) algorithm to classify audit findings based on the clauses of the ISO 9001:2015 QMS Requirements international standard. In this paper, the author explored how the neural network can predict the correct clause of the standard according to text patterns of audit findings. Based on modelling results, the neural network has generated a Cross Entropy (CE) values of 6.39, 18.09, 18.09 and Percentage Error (PE) values of 21.83, 21.58, and 22.39 in training, testing, and validation environments, respectively. Moreover, the model has achieved a combined Classification Accuracy (CA) of 96%, as for which, based on the actual implementation, the model has accurately predicted 95% of the audit findings analyzed.
Bookmarks Related papers MentionsView impact
In the field of mobile robotics, path planning is one of the most widely-sought areas of interest... more In the field of mobile robotics, path planning is one of the most widely-sought areas of interest due to its nature of complexity, where such issue is also practically evident in the case of mobile robots used for waste disposal purposes. To overcome issues on path planning, researchers have studied various classical and heuristic methods, however, the extent of optimization applicability and accuracy still remain an opportunity for further improvements. This paper presents the exploration of Artificial Neural Networks (ANN) in characterizing the path planning capability of a mobile waste-robot in order to improve navigational accuracy and path tracking time. The author utilized proximity and sound sensors as input vectors, dual H-bridge Direct Current (DC) motors as target vectors, and trained the ANN model using Levenberg-Marquardt (LM) and Scaled Conjugate (SCG) algorithms. Results revealed that LM was significantly more accurate than SCG algorithm in local path planning with Mean Square Error (MSE) values of 1.75966, 2.67946, and 2.04963, and Regression (R) values of 0.995671, 0.991247, and 0.983187 in training, testing, and validation environments, respectively. Furthermore, based on simulation results, LM was also found to be more accurate and faster than SCG with Pearson R correlation coefficients of r x =.975, n x =6, p x =0.001 and r y =.987, n y =6, p y =0.000 and path tracking time of 8.47s.
Bookmarks Related papers MentionsView impact
Uploads
Papers by Ralph Sherwin Corpuz