Most recent issue published online in the International Journal of Web and Grid Services.
International Journal of Web and Grid Services
http://www.inderscience.com/browse/index.php?journalID=47&year=2024&vol=20&issue=1
Inderscience Publishers Ltd
en-uk
support@inderscience.com
International Journal of Web and Grid Services
1741-1106
1741-1114
© 2024 Inderscience Enterprises Ltd.
© 2024 Inderscience Publishers Ltd
editor@inderscience.com
International Journal of Web and Grid Services
https://www.inderscience.com/images/files/coverImgs/ijwgs_scoverijwgs.jpg
http://www.inderscience.com/browse/index.php?journalID=47&year=2024&vol=20&issue=1
-
Effectively learn how to learn: a novel few-shot learning with meta-gradient memory
http://www.inderscience.com/link.php?id=137549
Recently, the importance of few-shot learning has tremendously grown due to its widespread applicability. Via few-shot learning, users can train their models with few data and maintain high generalisation ability. Meta-learning and continual learning models have demonstrated elegant performance in model development. However, unstable performance and catastrophic forgetting are still two fatal issues with regard to retaining the memory of knowledge about previous tasks when facing new tasks. In this paper, a novel method, enhanced model-agnostic meta-learning (EN-MAML), is proposed for blending the flexible adaptation characteristics of meta-learning and the stable performance of continual learning to tackle the above problems. Based on the proposed learning method, users can efficiently and effectively train the model in a stable manner with few data. Experiments show that when following the N-way K-shot experimental protocol, EN-MAML has higher accuracy, more stable performance and faster convergence than other state-of-the-art models on several real datasets.
Effectively learn how to learn: a novel few-shot learning with meta-gradient memory
Lin Hui; Yi-Cheng Chen
International Journal of Web and Grid Services, Vol. 20, No. 1 (2024) pp. 3 - 24
Recently, the importance of few-shot learning has tremendously grown due to its widespread applicability. Via few-shot learning, users can train their models with few data and maintain high generalisation ability. Meta-learning and continual learning models have demonstrated elegant performance in model development. However, unstable performance and catastrophic forgetting are still two fatal issues with regard to retaining the memory of knowledge about previous tasks when facing new tasks. In this paper, a novel method, enhanced model-agnostic meta-learning (EN-MAML), is proposed for blending the flexible adaptation characteristics of meta-learning and the stable performance of continual learning to tackle the above problems. Based on the proposed learning method, users can efficiently and effectively train the model in a stable manner with few data. Experiments show that when following the N-way K-shot experimental protocol, EN-MAML has higher accuracy, more stable performance and faster convergence than other state-of-the-art models on several real datasets.]]>
10.1504/IJWGS.2024.137549
International Journal of Web and Grid Services, Vol. 20, No. 1 (2024) pp. 3 - 24
Lin Hui
Yi-Cheng Chen
Department of Computer Science and Information Engineering, Tamkang University, Taiwan ' Department of Information Management, National Central University, Taiwan
machine learning
deep learning
meta-learning
continual learning
2024-03-25T23:20:50-05:00
Copyright © 2024 Inderscience Enterprises Ltd.
20
1
3
24
2024-03-25T23:20:50-05:00
-
Smart contracts and marketplace for just-in-time management of pharmaceutical drugs
http://www.inderscience.com/link.php?id=137553
Blockchain technology has recently been used to provide a secure storage environment through a distributed ledger. Blockchain has increasingly been used in other sectors such as real estate and supply chains, where trust and transparency are paramount considerations. In the pharmaceutical industry, for operational efficiencies, information must be shared reliably between the various stakeholders. A significant limitation in the existing literature is the lack of work to address niche problems such as the just-in-time disposal of drugs that are close to expiry. To address this gap, we propose using blockchain technology. The architectural underpinning of the proposed system (PharmaBlock) is presented and discussed. The primary contribution of this paper is the use of an early warning system (EWS) coupled with marketplace to intelligently identify and dispose of near-expiry drugs. The EWS and marketplace are evaluated and benchmarked using an experimental setup. The result of this experimental has shown that over 90% of notifications were sent correctly and shown also more than 92% of the optimal prices were predicted correctly in PharmaBlock.
Smart contracts and marketplace for just-in-time management of pharmaceutical drugs
Abeer Rashad Mirdad; Abdulaziz Mohammed Khan; Farookh Khadeer Hussain
International Journal of Web and Grid Services, Vol. 20, No. 1 (2024) pp. 25 - 53
Blockchain technology has recently been used to provide a secure storage environment through a distributed ledger. Blockchain has increasingly been used in other sectors such as real estate and supply chains, where trust and transparency are paramount considerations. In the pharmaceutical industry, for operational efficiencies, information must be shared reliably between the various stakeholders. A significant limitation in the existing literature is the lack of work to address niche problems such as the just-in-time disposal of drugs that are close to expiry. To address this gap, we propose using blockchain technology. The architectural underpinning of the proposed system (PharmaBlock) is presented and discussed. The primary contribution of this paper is the use of an early warning system (EWS) coupled with marketplace to intelligently identify and dispose of near-expiry drugs. The EWS and marketplace are evaluated and benchmarked using an experimental setup. The result of this experimental has shown that over 90% of notifications were sent correctly and shown also more than 92% of the optimal prices were predicted correctly in PharmaBlock.]]>
10.1504/IJWGS.2024.137553
International Journal of Web and Grid Services, Vol. 20, No. 1 (2024) pp. 25 - 53
Abeer Rashad Mirdad
Abdulaziz Mohammed Khan
Farookh Khadeer Hussain
School of Computer Science, Australian Artificial Intelligence Institute (AAII), Faculty of Engineering and Information Technology, University of Technology Sydney, Sydney, 15 Broadway, Ultimo, NSW 2007, Australia ' School of Computer Science, Australian Artificial Intelligence Institute (AAII), Faculty of Engineering and Information Technology, University of Technology Sydney, Sydney, 15 Broadway, Ultimo, NSW 2007, Australia ' School of Computer Science, Australian Artificial Intelligence Institute (AAII), Faculty of Engineering and Information Technology, University of Technology Sydney, Sydney, 15 Broadway, Ultimo, NSW 2007, Australia
blockchain
smart contracts
early warning system
e-marketplace
pharmaceutical supply chain
research challenges
2024-03-25T23:20:50-05:00
Copyright © 2024 Inderscience Enterprises Ltd.
20
1
25
53
2024-03-25T23:20:50-05:00
-
An augmented interpretive framework based on aspect sentiment words aggregation
http://www.inderscience.com/link.php?id=137551
Given the mounting anxieties surrounding the interpretability of neural models, appraising interpretability remains an unsolved puzzle owing to the ineffectual performance of existing interpretation techniques and evaluation metrics. The architecture of neural network models varies depending on the task at hand, making it challenging to devise a universal method of explanation that can produce coherent justifications for each model. This paper proposes a framework to enhance the interpretability of text sentiment classification models using aspect sentiment words (ASW) aggregation, which can be applied to web services to improve transparency, accountability, and user trust. The proposed method extracts ASW from sentences and consolidates the token importance scores to provide more credible justifications. The paper also introduces new evaluation metrics for faithfulness, which assess whether interpretations accurately reflect the model's decision-making process. The proposed metrics are effective in evaluating the fidelity of rationales to models at the snippet-level.
An augmented interpretive framework based on aspect sentiment words aggregation
Chao Li; Bo Shen; Yingsi Zhao; Qing-An Zeng
International Journal of Web and Grid Services, Vol. 20, No. 1 (2024) pp. 54 - 73
Given the mounting anxieties surrounding the interpretability of neural models, appraising interpretability remains an unsolved puzzle owing to the ineffectual performance of existing interpretation techniques and evaluation metrics. The architecture of neural network models varies depending on the task at hand, making it challenging to devise a universal method of explanation that can produce coherent justifications for each model. This paper proposes a framework to enhance the interpretability of text sentiment classification models using aspect sentiment words (ASW) aggregation, which can be applied to web services to improve transparency, accountability, and user trust. The proposed method extracts ASW from sentences and consolidates the token importance scores to provide more credible justifications. The paper also introduces new evaluation metrics for faithfulness, which assess whether interpretations accurately reflect the model's decision-making process. The proposed metrics are effective in evaluating the fidelity of rationales to models at the snippet-level.]]>
10.1504/IJWGS.2024.137551
International Journal of Web and Grid Services, Vol. 20, No. 1 (2024) pp. 54 - 73
Chao Li
Bo Shen
Yingsi Zhao
Qing-An Zeng
School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing, 100044, China ' Key Laboratory of Communication and Information Systems, Beijing Jiaotong University, Beijing, 100044, China ' School of Economics and Management, Beijing Jiaotong University, Beijing, 100044, China ' Department of Computer Systems Technology, North Carolina A&T State University, North Carolina, 27695, USA
deep learning
text sentiment classification
interpretability
aspect sentiment words aggregation
2024-03-25T23:20:50-05:00
Copyright © 2024 Inderscience Enterprises Ltd.
20
1
54
73
2024-03-25T23:20:50-05:00
-
Attention-based mechanism and feature fusion network for person re-identification
http://www.inderscience.com/link.php?id=137566
For the problem that person features cannot be sufficiently extracted in person re-identification, a person re-identification model based on attention mechanism is proposed. Firstly, person features are extracted using a hybrid network combining transformer's core multi-headed self-attentive module with the convolutional neural network ResNet50-IBN-a. Secondly, an efficient channel attention mechanism ECANet is embedded to make the model of this paper more focused on the key information in the person foreground. Finally, fusing the mid-level and high-level features in the model can avoid some discriminative features loss. The experimental results show that the provide model achieves 94.8% rank-1 and 84.5% mAP on the Market-1501 dataset; achieves 84.9% rank-1 and 65.9% mAP on the DukeMTMC-reID dataset; and achieves 40.3% rank-1 and 33.3% mAP on the Occluded-Duke MTMC dataset. Our proposed model compares well with some of the existing person re-identification models on these datasets mentioned above.
Attention-based mechanism and feature fusion network for person re-identification
Mingshou An; Yunchuan He; Hye-Youn Lim; Dae-Seong Kang
International Journal of Web and Grid Services, Vol. 20, No. 1 (2024) pp. 74 - 92
For the problem that person features cannot be sufficiently extracted in person re-identification, a person re-identification model based on attention mechanism is proposed. Firstly, person features are extracted using a hybrid network combining transformer's core multi-headed self-attentive module with the convolutional neural network ResNet50-IBN-a. Secondly, an efficient channel attention mechanism ECANet is embedded to make the model of this paper more focused on the key information in the person foreground. Finally, fusing the mid-level and high-level features in the model can avoid some discriminative features loss. The experimental results show that the provide model achieves 94.8% rank-1 and 84.5% mAP on the Market-1501 dataset; achieves 84.9% rank-1 and 65.9% mAP on the DukeMTMC-reID dataset; and achieves 40.3% rank-1 and 33.3% mAP on the Occluded-Duke MTMC dataset. Our proposed model compares well with some of the existing person re-identification models on these datasets mentioned above.]]>
10.1504/IJWGS.2024.137566
International Journal of Web and Grid Services, Vol. 20, No. 1 (2024) pp. 74 - 92
Mingshou An
Yunchuan He
Hye-Youn Lim
Dae-Seong Kang
School of Computer Science and Engineering, Xi'an Technological University, Xi'an, China ' School of Computer Science and Engineering, Xi'an Technological University, Xi'an, China ' Department of Electronics Engineering, Dong-A University, Busan, South Korea ' Department of Electronics Engineering, Dong-A University, Busan, South Korea
attention mechanism
person re-identification
feature fusion
convolutional neural network
occluded person detection
2024-03-25T23:20:50-05:00
Copyright © 2024 Inderscience Enterprises Ltd.
20
1
74
92
2024-03-25T23:20:50-05:00
-
Simplified swarm optimisation for CNN hyperparameters: a sound classification approach
http://www.inderscience.com/link.php?id=137557
The pervasive integration of environmental sounds into diverse aspects of daily life - ranging from smart city management, accurate location pinpointing, surveillance mechanisms, auditory machine functionalities, to environmental monitoring - is evident. Central to this is environmental sound classification, gaining academic traction. However, sound classifications present challenges due to the variables causing noise. This research aimed to discern the convolutional neural network (CNN) model with optimal accuracy in ESC tasks via hyperparameter optimisation. Simplified swarm optimisation (SSO) algorithm was harnessed to encapsulate the CNN architecture, providing an untransformed representation of CNN hyperparameters during optimisation. Utilising the prominent datasets and applying data augmentation techniques, the CNN model designed via SSO achieved accuracies of 99.01%, 97.42%, and 98.96% respectively. Compared to prior studies, this denotes the highest accuracy from a pure CNN model, advancing automated CNN design for urban sound classification.
Simplified swarm optimisation for CNN hyperparameters: a sound classification approach
Zhenyao Liu; Wei-Chang Yeh
International Journal of Web and Grid Services, Vol. 20, No. 1 (2024) pp. 93 - 113
The pervasive integration of environmental sounds into diverse aspects of daily life - ranging from smart city management, accurate location pinpointing, surveillance mechanisms, auditory machine functionalities, to environmental monitoring - is evident. Central to this is environmental sound classification, gaining academic traction. However, sound classifications present challenges due to the variables causing noise. This research aimed to discern the convolutional neural network (CNN) model with optimal accuracy in ESC tasks via hyperparameter optimisation. Simplified swarm optimisation (SSO) algorithm was harnessed to encapsulate the CNN architecture, providing an untransformed representation of CNN hyperparameters during optimisation. Utilising the prominent datasets and applying data augmentation techniques, the CNN model designed via SSO achieved accuracies of 99.01%, 97.42%, and 98.96% respectively. Compared to prior studies, this denotes the highest accuracy from a pure CNN model, advancing automated CNN design for urban sound classification.]]>
10.1504/IJWGS.2024.137557
International Journal of Web and Grid Services, Vol. 20, No. 1 (2024) pp. 93 - 113
Zhenyao Liu
Wei-Chang Yeh
Integration and Collaboration Laboratory, Department of Industrial Engineering and Management Engineering, National Tsing Hua University, Hsinchu, Taiwan ' Integration and Collaboration Laboratory, Department of Industrial Engineering and Management Engineering, National Tsing Hua University, Hsinchu, Taiwan
convolutional neural network
CNN
simplified swarm optimisation
SSO
environmental sound classification
ESC
hyperparameter optimisation
2024-03-25T23:20:50-05:00
Copyright © 2024 Inderscience Enterprises Ltd.
20
1
93
113
2024-03-25T23:20:50-05:00
-
C3D-LSTM: a novel convolution-3D-based LSTM for link prediction in dynamic social networks
http://www.inderscience.com/link.php?id=137563
Recently, due to the surge in the use of social networks, link prediction has become an essential technique which could enable service providers to anticipate future friendships between users based on the network structure and personal data so as to enhance consumer loyalty and experience. Undoubtedly, link prediction analysis becomes increasingly difficult when social networks expand quickly, particularly in light of the major advancements in complex social network modelling. Prior studies which predicted social links based on static network settings may have ignored the dynamic variation of networks over time. In this research, an end-to-end model, convolution-3D-based long-short-term memory (abbreviated as C3D-LSTM), is developed to integrate the convolution neural network (CNN) and long-short-term memory (LSTM) network for effective link prediction. We employ 3D convolution to detect subtle patterns in social network snapshots, capturing short-term spatial-temporal features. LSTM layers then interpret these features to model the network's long-term temporal dynamics. To demonstrate its practicability, extensive experiments are conducted to show that C3D-LSTM surpasses current state-of-the-art techniques and delivers remarkable performance.
C3D-LSTM: a novel convolution-3D-based LSTM for link prediction in dynamic social networks
Yi-Cheng Chen; Tipajin Thaipisutikul
International Journal of Web and Grid Services, Vol. 20, No. 1 (2024) pp. 114 - 134
Recently, due to the surge in the use of social networks, link prediction has become an essential technique which could enable service providers to anticipate future friendships between users based on the network structure and personal data so as to enhance consumer loyalty and experience. Undoubtedly, link prediction analysis becomes increasingly difficult when social networks expand quickly, particularly in light of the major advancements in complex social network modelling. Prior studies which predicted social links based on static network settings may have ignored the dynamic variation of networks over time. In this research, an end-to-end model, convolution-3D-based long-short-term memory (abbreviated as C3D-LSTM), is developed to integrate the convolution neural network (CNN) and long-short-term memory (LSTM) network for effective link prediction. We employ 3D convolution to detect subtle patterns in social network snapshots, capturing short-term spatial-temporal features. LSTM layers then interpret these features to model the network's long-term temporal dynamics. To demonstrate its practicability, extensive experiments are conducted to show that C3D-LSTM surpasses current state-of-the-art techniques and delivers remarkable performance.]]>
10.1504/IJWGS.2024.137563
International Journal of Web and Grid Services, Vol. 20, No. 1 (2024) pp. 114 - 134
Yi-Cheng Chen
Tipajin Thaipisutikul
Department of Information Management, National Central University, Taiwan ' Faculty of Information and Communication Technology, Mahidol University, Nakhon Pathom, Thailand
deep learning
convolution neural network
CNN
link prediction
long-short-term memory network
LSTM
social network
2024-03-25T23:20:50-05:00
Copyright © 2024 Inderscience Enterprises Ltd.
20
1
114
134
2024-03-25T23:20:50-05:00