Pin Shen Teh
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Pin Shen Teh has a total of 45 co-authors. Here are the top co-authors by number of shared publications:
- Andrew Beng Jin Teoh
School of Electrical and Electronic Engineering, College of Engineering, Yonsei University, Seoul, Republic of Korea
10 shared publications - Raheel Nawaz
Staffordshire University, Stoke-on-Trent, United Kingdom
6 shared publications Ning Zhang
University of Manchester
4 shared publications- Raheem Sarwar
OTEHM, Manchester Metropolitan University, Manchester, UK
4 shared publications - Ke Chen
Zhejiang University
3 shared publications - Muhammad Umair Hassan
Department of ICT and Natural Sciences, Norwegian University of Science and Technology (NTNU), 6009 Ă…lesund, Norway
3 shared publications
Latest Publications
- Open Access
- 14 June 2026
- Published by Multimedia University in Journal of Informatics and Web Engineering
- Vol. 5  (2) , 250-265
AI-Driven Intrusion Detection System for Network Security
In the evolving landscape of cybersecurity, Intrusion Detection Systems (IDS) play a vital role in safeguarding computer networks against malicious activity. Traditional signature-based IDS approaches are increasingly ineffective in detecting novel, complex, or zero-day attacks due to their reliance on predefined rules. To overcome these limitations, this study proposes an AI-driven IDS that integrates both classical machine learning and modern deep learning techniques. The framework introduces and compares three models such as Support Vector Machine (SVM), Convolutional Neural Network (CNN), and a hybrid CNN-SVM model. The system is designed to analyze network traffic patterns and classify them as either benign or malicious, enhancing detection capabilities through intelligent feature learning and classification. Two widely recognized benchmark datasets are used to train and validate the models, ensuring the system's applicability to a variety of network environments. Furthermore, the research includes the development of a real-time detection component that incorporates live packet capture, feature extraction, and dynamic visualization via a dashboard interface. This paper contributes to the field by demonstrating how hybrid AI models can effectively address the challenges of network intrusion detection. The study emphasizes the importance of combining traditional and deep learning approaches to build scalable, adaptive, and accurate intrusion detection systems for modern network infrastructures.
- 14 June 2026
- Published by Multimedia University in Journal of Informatics and Web Engineering
- Vol. 5  (2) , 250-265
Open AccessAI-Driven Intrusion Detection System for Network Security
In the evolving landscape of cybersecurity, Intrusion Detection Systems (IDS) play a vital role in safeguarding computer networks against malicious activity. Traditional signature-based IDS approaches are increasingly ineffective in detecting novel, complex, or zero-day attacks due to their reliance on predefined rules. To overcome these limitations, this study proposes an AI-driven IDS that integrates both classical machine learning and modern deep learning techniques. The framework introduces and compares three models such as Support Vector Machine (SVM), Convolutional Neural Network (CNN), and a hybrid CNN-SVM model. The system is designed to analyze network traffic patterns and classify them as either benign or malicious, enhancing detection capabilities through intelligent feature learning and classification. Two widely recognized benchmark datasets are used to train and validate the models, ensuring the system's applicability to a variety of network environments. Furthermore, the research includes the development of a real-time detection component that incorporates live packet capture, feature extraction, and dynamic visualization via a dashboard interface. This paper contributes to the field by demonstrating how hybrid AI models can effectively address the challenges of network intrusion detection. The study emphasizes the importance of combining traditional and deep learning approaches to build scalable, adaptive, and accurate intrusion detection systems for modern network infrastructures.
- Open Access
- 16 October 2024
- Published by Multimedia University in Journal of Informatics and Web Engineering
- Vol. 3  (3) , 233-255
HybridEval: An Improved Novel Hybrid Metric for Evaluation of Text Summarization
The present work re-evaluates the evaluation method for text summarization tasks. Two state-of-the-art assessment measures e.g., Recall-Oriented Understudy for Gisting Evaluation (ROUGE) and Bilingual Evaluation Understudy (BLEU) are discussed along with their limitations before presenting a novel evaluation metric. The evaluation scores are significantly different because of the length and vocabulary of the sentences, this suggests that the primary restriction is its inability to preserve the semantics and meaning of the sentences and consistent weight distribution over the whole sentence. To address this, the present work organizes the phrases into six different groups and to evaluate “text summarization” problems, a new hybrid approach (HybridEval) is proposed. Our approach uses a weighted sum of cosine scores from InferSent’s SentEval algorithms combined with original scores, achieving high accuracy. HybridEval outperforms existing state-of-the-art models by 10-15% in evaluation scores.
- 16 October 2024
- Published by Multimedia University in Journal of Informatics and Web Engineering
- Vol. 3  (3) , 233-255
Open AccessHybridEval: An Improved Novel Hybrid Metric for Evaluation of Text Summarization
The present work re-evaluates the evaluation method for text summarization tasks. Two state-of-the-art assessment measures e.g., Recall-Oriented Understudy for Gisting Evaluation (ROUGE) and Bilingual Evaluation Understudy (BLEU) are discussed along with their limitations before presenting a novel evaluation metric. The evaluation scores are significantly different because of the length and vocabulary of the sentences, this suggests that the primary restriction is its inability to preserve the semantics and meaning of the sentences and consistent weight distribution over the whole sentence. To address this, the present work organizes the phrases into six different groups and to evaluate “text summarization” problems, a new hybrid approach (HybridEval) is proposed. Our approach uses a weighted sum of cosine scores from InferSent’s SentEval algorithms combined with original scores, achieving high accuracy. HybridEval outperforms existing state-of-the-art models by 10-15% in evaluation scores.
- Open Access
- 16 October 2024
- Published by Multimedia University in Journal of Informatics and Web Engineering
- Vol. 3  (3) , 256-270
Intelligent Abstractive Summarization of Scholarly Publications with Transfer Learning
Intelligent abstractive text summarization of scholarly publications refers to machine-generated summaries that capture the essential ideas of an article while maintaining semantic coherence and grammatical accuracy. As information continues to grow at an overwhelming rate, text summarization has emerged as a critical area of research. In the past, summarization of scientific publications predominantly relied on extractive methods. These approaches involve selecting key sentences or phrases directly from the original document to create a summary or generate a suitable title. Although extractive methods preserve the original wording, they often lack the ability to produce a coherent, concise, and fluent summary, especially when dealing with complex or lengthy texts. In contrast, abstractive summarization represents a more sophisticated approach. Rather than extracting content from the source, abstractive models generate summaries using new language, often incorporating words and phrases not found in the original text. This allows for more natural, human-like summaries that better capture the key ideas in a fluid and cohesive manner. This study introduces two advanced models for generating titles from the abstracts of scientific articles. The first model employs a Gated Recurrent Unit (GRU) encoder coupled with a greedy-search decoder, while the second utilizes a Transformer model, known for its capacity to handle long-range dependencies in text. The findings demonstrate that both models outperform the baseline Long Short-Term Memory (LSTM) model in terms of efficiency and fluency. Specifically, the GRU model achieved a ROUGE-1 score of 0.2336, and the Transformer model scored 0.2881, significantly higher than the baseline LSTM model, which reported a ROUGE-1 score of 0.1033. These results underscore the potential of abstractive models to enhance the quality and accuracy of summarization in academic and scholarly contexts, offering more intuitive and meaningful summaries.
- 16 October 2024
- Published by Multimedia University in Journal of Informatics and Web Engineering
- Vol. 3  (3) , 256-270
Open AccessIntelligent Abstractive Summarization of Scholarly Publications with Transfer Learning
Intelligent abstractive text summarization of scholarly publications refers to machine-generated summaries that capture the essential ideas of an article while maintaining semantic coherence and grammatical accuracy. As information continues to grow at an overwhelming rate, text summarization has emerged as a critical area of research. In the past, summarization of scientific publications predominantly relied on extractive methods. These approaches involve selecting key sentences or phrases directly from the original document to create a summary or generate a suitable title. Although extractive methods preserve the original wording, they often lack the ability to produce a coherent, concise, and fluent summary, especially when dealing with complex or lengthy texts. In contrast, abstractive summarization represents a more sophisticated approach. Rather than extracting content from the source, abstractive models generate summaries using new language, often incorporating words and phrases not found in the original text. This allows for more natural, human-like summaries that better capture the key ideas in a fluid and cohesive manner. This study introduces two advanced models for generating titles from the abstracts of scientific articles. The first model employs a Gated Recurrent Unit (GRU) encoder coupled with a greedy-search decoder, while the second utilizes a Transformer model, known for its capacity to handle long-range dependencies in text. The findings demonstrate that both models outperform the baseline Long Short-Term Memory (LSTM) model in terms of efficiency and fluency. Specifically, the GRU model achieved a ROUGE-1 score of 0.2336, and the Transformer model scored 0.2881, significantly higher than the baseline LSTM model, which reported a ROUGE-1 score of 0.1033. These results underscore the potential of abstractive models to enhance the quality and accuracy of summarization in academic and scholarly contexts, offering more intuitive and meaningful summaries.
- Open Access
- 31 May 2024
- Published by Universitas Ahmad Dahlan in International Journal of Advances in Intelligent Informatics
- Vol. 10  (2) , 202-219
Abnormal behavior recognition using SRU with attention mechanism
In response to the critical need for enhanced public safety measures, this study introduces an advanced intelligent surveillance system designed to autonomously detect abnormal behaviors within public spaces. Leveraging the computational efficiency and accuracy of a Simple Recurrent Unit (SRU) integrated with an attention mechanism, this research delivers a novel approach towards understanding and interpreting human interactions in real-time video footage. Distinctively, the model specializes in identifying two primary categories of abnormal behavior: aggressive two-person interactions such as physical confrontations and collective crowd dynamics, characterized by sudden dispersal patterns indicative of distress or danger. The incorporation of Attention mechanism precisely targets critical elements of behavior, thereby enhancing the model's focus and interpretative clarity. Empirical validation across five benchmark datasets reveals that our model not only outperforms traditional Long Short-Term Memory (LSTM) frameworks in terms of speed by a factor of 1.5 but also demonstrates superior accuracy in abnormal behavior recognition. These findings not only underscore the model's potential in preempting potential safety threats but also mark a significant advancement in the application of deep learning technologies for public security infrastructures. This research contributes to the broader discourse on public safety, offering actionable insights and robust technological solutions to enhance surveillance efficacy and response mechanisms in critical public domains.
- 31 May 2024
- Published by Universitas Ahmad Dahlan in International Journal of Advances in Intelligent Informatics
- Vol. 10  (2) , 202-219
Open AccessAbnormal behavior recognition using SRU with attention mechanism
In response to the critical need for enhanced public safety measures, this study introduces an advanced intelligent surveillance system designed to autonomously detect abnormal behaviors within public spaces. Leveraging the computational efficiency and accuracy of a Simple Recurrent Unit (SRU) integrated with an attention mechanism, this research delivers a novel approach towards understanding and interpreting human interactions in real-time video footage. Distinctively, the model specializes in identifying two primary categories of abnormal behavior: aggressive two-person interactions such as physical confrontations and collective crowd dynamics, characterized by sudden dispersal patterns indicative of distress or danger. The incorporation of Attention mechanism precisely targets critical elements of behavior, thereby enhancing the model's focus and interpretative clarity. Empirical validation across five benchmark datasets reveals that our model not only outperforms traditional Long Short-Term Memory (LSTM) frameworks in terms of speed by a factor of 1.5 but also demonstrates superior accuracy in abnormal behavior recognition. These findings not only underscore the model's potential in preempting potential safety threats but also mark a significant advancement in the application of deep learning technologies for public security infrastructures. This research contributes to the broader discourse on public safety, offering actionable insights and robust technological solutions to enhance surveillance efficacy and response mechanisms in critical public domains.
- Open Access
- 10 May 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Asian and Low-Resource Language Information Processing
- Vol. 23  (5) , 1-14
Crossing Linguistic Barriers: Authorship Attribution in Sinhala Texts
Authorship attribution involves determining the original author of an anonymous text from a pool of potential authors. The author attribution task has applications in several domains, such as plagiarism detection, digital text forensics, and information retrieval. While these applications extend beyond any single language, existing research has predominantly centered on English, posing challenges for application in languages such as Sinhala due to linguistic disparities and a lack of language processing tools. We present the first comprehensive study on cross-topic authorship attribution for Sinhala texts and propose a solution that can effectively perform the authorship attribution task even if the topics within the test and training samples differ. Our solution consists of three main parts: (i) extraction of topic-independent stylometric features, (ii) generation of a small candidate author set with the help of similarity search, and (iii) identification of the true author. Several experimental studies were carried out to demonstrate that the proposed solution can effectively handle real-world scenarios involving a large number of candidate authors and a limited number of text samples for each candidate author.
- 10 May 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Asian and Low-Resource Language Information Processing
- Vol. 23  (5) , 1-14
Open AccessCrossing Linguistic Barriers: Authorship Attribution in Sinhala Texts
Authorship attribution involves determining the original author of an anonymous text from a pool of potential authors. The author attribution task has applications in several domains, such as plagiarism detection, digital text forensics, and information retrieval. While these applications extend beyond any single language, existing research has predominantly centered on English, posing challenges for application in languages such as Sinhala due to linguistic disparities and a lack of language processing tools. We present the first comprehensive study on cross-topic authorship attribution for Sinhala texts and propose a solution that can effectively perform the authorship attribution task even if the topics within the test and training samples differ. Our solution consists of three main parts: (i) extraction of topic-independent stylometric features, (ii) generation of a small candidate author set with the help of similarity search, and (iii) identification of the true author. Several experimental studies were carried out to demonstrate that the proposed solution can effectively handle real-world scenarios involving a large number of candidate authors and a limited number of text samples for each candidate author.
- Open Access
- 24 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 15399-15409
AGI-P: A Gender Identification Framework for Authorship Analysis Using Customized Fine-Tuning of Multilingual Language Model
In this investigation, we propose a solution for the author’s gender identification task called AGI-P. This task has several real-world applications across different fields, such as marketing and advertising, forensic linguistics, sociology, recommendation systems, language processing, historical analysis, education, and language learning. We created a new dataset to evaluate our proposed method. The dataset is balanced in terms of gender using a random sampling method and consists of 1944 samples in total. We use accuracy as an evaluation measure and compare the performance of the proposed solution (AGI-P) against state-of-the-art machine learning classifiers and fine-tuned pre-trained multilingual language models such as DistilBERT, mBERT, XLM-RoBERTa, and Multilingual DEBERTa. In this regard, we also propose a customized fine-tuning strategy that improves the accuracy of the pre-trained language models for the author gender identification task. Our extensive experimental studies reveal that our solution (AGI-P) outperforms the well-known machine learning classifiers and fine-tuned pre-trained multilingual language models with an accuracy level of 92.03%. Moreover, the pre-trained multilingual language models, fine-tuned with the proposed customized strategy, outperform the fine-tuned pre-trained language models using an out-of-the-box fine-tuning strategy. The codebase and corpus can be accessed on our GitHub page at: https://github.com/mumairhassan/AGI-P
- 24 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 15399-15409
Open AccessAGI-P: A Gender Identification Framework for Authorship Analysis Using Customized Fine-Tuning of Multilingual Language Model
In this investigation, we propose a solution for the author’s gender identification task called AGI-P. This task has several real-world applications across different fields, such as marketing and advertising, forensic linguistics, sociology, recommendation systems, language processing, historical analysis, education, and language learning. We created a new dataset to evaluate our proposed method. The dataset is balanced in terms of gender using a random sampling method and consists of 1944 samples in total. We use accuracy as an evaluation measure and compare the performance of the proposed solution (AGI-P) against state-of-the-art machine learning classifiers and fine-tuned pre-trained multilingual language models such as DistilBERT, mBERT, XLM-RoBERTa, and Multilingual DEBERTa. In this regard, we also propose a customized fine-tuning strategy that improves the accuracy of the pre-trained language models for the author gender identification task. Our extensive experimental studies reveal that our solution (AGI-P) outperforms the well-known machine learning classifiers and fine-tuned pre-trained multilingual language models with an accuracy level of 92.03%. Moreover, the pre-trained multilingual language models, fine-tuned with the proposed customized strategy, outperform the fine-tuned pre-trained language models using an out-of-the-box fine-tuning strategy. The codebase and corpus can be accessed on our GitHub page at: https://github.com/mumairhassan/AGI-P
- 15 March 2023
- Published by Springer Nature
- p. 465-476
Minecraft as a Tool to Enhance Engagement in Higher Education
- 15 March 2023
- Published by Springer Nature
- p. 465-476
Minecraft as a Tool to Enhance Engagement in Higher Education
- Open Access
- 9 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 5069-5088
A Review of Abnormal Behavior Detection in Activities of Daily Living
Abnormal behavior detection (ABD) systems are built to automatically identify and recognize abnormal behavior from various input data types, such as sensor-based and vision-based input. As much as the attention received for ABD systems, the number of studies on ABD in activities of daily living (ADL) is limited. Owing to the increasing rate of elderly accidents in the home compound, ABD in ADL research should be given as much attention to preventing accidents by sending out signals when abnormal behavior such as falling is detected. In this study, we compare and contrast the formation of the ABD system in ADL from input data types (sensor-based input and vision-based input) to modeling techniques (conventional and deep learning approaches). We scrutinize the public datasets available and provide solutions for one of the significant issues: the lack of datasets in ABD in ADL. This work aims to guide new research to understand the field of ABD in ADL better and serve as a reference for future study of better Ambient Assisted Living with the growing smart home trend.
- 9 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 5069-5088
Open AccessA Review of Abnormal Behavior Detection in Activities of Daily Living
Abnormal behavior detection (ABD) systems are built to automatically identify and recognize abnormal behavior from various input data types, such as sensor-based and vision-based input. As much as the attention received for ABD systems, the number of studies on ABD in activities of daily living (ADL) is limited. Owing to the increasing rate of elderly accidents in the home compound, ABD in ADL research should be given as much attention to preventing accidents by sending out signals when abnormal behavior such as falling is detected. In this study, we compare and contrast the formation of the ABD system in ADL from input data types (sensor-based input and vision-based input) to modeling techniques (conventional and deep learning approaches). We scrutinize the public datasets available and provide solutions for one of the significant issues: the lack of datasets in ABD in ADL. This work aims to guide new research to understand the field of ABD in ADL better and serve as a reference for future study of better Ambient Assisted Living with the growing smart home trend.
- Open Access
- 24 December 2019
- Published by Springer Nature in Journal of Ambient Intelligence and Humanized Computing
- Vol. 11  (10) , 4019-4039
Strengthen user authentication on mobile devices by using user’s touch dynamics pattern
Mobile devices, particularly the touch screen mobile devices, are increasingly used to store and access private and sensitive data or services, and this has led to an increased demand for more secure and usable security services, one of which is user authentication. Currently, mobile device authentication services mainly use a knowledge-based method, e.g. a PIN-based authentication method, and, in some cases, a fingerprint-based authentication method is also supported. The knowledge-based method is vulnerable to impersonation attacks, while the fingerprint-based method can be unreliable sometimes. To overcome these limitations and to make the authentication service more secure and reliable for touch screen mobile device users, we have investigated the use of touch dynamics biometrics as a mobile device authentication solution by designing, implementing and evaluating a touch dynamics authentication method. This paper describes the design, implementation, and evaluation of this method, the acquisition of raw touch dynamics data, the use of the raw data to obtain touch dynamics features, and the training of the features to build an authentication model for user identity verification. The evaluation results show that by integrating the touch dynamics authentication method into the PIN-based authentication method, the protection levels against impersonation attacks is greatly enhanced. For example, if a PIN is compromised, the success rate of an impersonation attempt is drastically reduced from 100% (if only a 4-digit PIN is used) to 9.9% (if both the PIN and the touch dynamics are used).
- 24 December 2019
- Published by Springer Nature in Journal of Ambient Intelligence and Humanized Computing
- Vol. 11  (10) , 4019-4039
Open AccessStrengthen user authentication on mobile devices by using user’s touch dynamics pattern
Mobile devices, particularly the touch screen mobile devices, are increasingly used to store and access private and sensitive data or services, and this has led to an increased demand for more secure and usable security services, one of which is user authentication. Currently, mobile device authentication services mainly use a knowledge-based method, e.g. a PIN-based authentication method, and, in some cases, a fingerprint-based authentication method is also supported. The knowledge-based method is vulnerable to impersonation attacks, while the fingerprint-based method can be unreliable sometimes. To overcome these limitations and to make the authentication service more secure and reliable for touch screen mobile device users, we have investigated the use of touch dynamics biometrics as a mobile device authentication solution by designing, implementing and evaluating a touch dynamics authentication method. This paper describes the design, implementation, and evaluation of this method, the acquisition of raw touch dynamics data, the use of the raw data to obtain touch dynamics features, and the training of the features to build an authentication model for user identity verification. The evaluation results show that by integrating the touch dynamics authentication method into the PIN-based authentication method, the protection levels against impersonation attacks is greatly enhanced. For example, if a PIN is compromised, the success rate of an impersonation attempt is drastically reduced from 100% (if only a 4-digit PIN is used) to 9.9% (if both the PIN and the touch dynamics are used).
- 1 November 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Abnormal Behavior Recognition using CNN-LSTM with Attention Mechanism
There is a rising trend of security issues in our society nowadays. Every now and then, there are news such as robberies, fighting or terrorism around the world. Hence, some robust measurements need to be done to ensure public safety. This is when computer vision techniques come into play. Conventional surveillance cameras lack the capability of autonomously detecting abnormal behaviors in footages, and hence the determination of abnormal activities is solely dependent on human judgement. There is no absolute meaning of what abnormal behavior is, it depends on the settings. For example, fighting in a martial art class is a normal behavior, however if there is fighting in a bank, it is considered as abnormal behavior. In this study, we focus on two scopes: two-persons interactions and crowd-based interactions. Our Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) with attention mechanism model can automatically extract the important features from the video frames and interpret the temporal information between the video sequences. Different from the typical neural networks, our model includes attention mechanism that focuses on salient part of human action. Five benchmark datasets are used to validate the performance of the proposed model.
- 1 November 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Abnormal Behavior Recognition using CNN-LSTM with Attention Mechanism
There is a rising trend of security issues in our society nowadays. Every now and then, there are news such as robberies, fighting or terrorism around the world. Hence, some robust measurements need to be done to ensure public safety. This is when computer vision techniques come into play. Conventional surveillance cameras lack the capability of autonomously detecting abnormal behaviors in footages, and hence the determination of abnormal activities is solely dependent on human judgement. There is no absolute meaning of what abnormal behavior is, it depends on the settings. For example, fighting in a martial art class is a normal behavior, however if there is fighting in a bank, it is considered as abnormal behavior. In this study, we focus on two scopes: two-persons interactions and crowd-based interactions. Our Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) with attention mechanism model can automatically extract the important features from the video frames and interpret the temporal information between the video sequences. Different from the typical neural networks, our model includes attention mechanism that focuses on salient part of human action. Five benchmark datasets are used to validate the performance of the proposed model.