International Journal of Web Information Systems

Name: International Journal of Web Information Systems
Creator: vLex
License: https://vlex.com/terms-of-service/

Publisher:: Emerald Group Publishing Limited
Publication date:: 2021-02-01

ISBN:: 1744-0084

Issue Number

No. 21-1, November - December 2024

Latest documents

Conv-DMSA: an efficient imputation model for multivariate time series through diagonal mask self-attention
Purpose: The proposed model aims to tackle the data quality issues in multivariate time series caused by missing values. It preserves data set integrity by accurately imputing missing data, ensuring reliable analysis outcomes. Design/methodology/approach: The Conv-DMSA model employs a combination of self-attention mechanisms and convolutional networks to handle the complexities of multivariate time series data. The convolutional network is adept at learning features across uneven time intervals through an imputation feature map, while the Diagonal Mask Self-Attention (DMSA) block is specifically designed to capture time dependencies and feature correlations. This dual approach allows the model to effectively address the temporal imbalance, feature correlation and time dependency challenges that are often overlooked in traditional imputation models. Findings: Extensive experiments conducted on two public data sets and a real project data set have demonstrated the adaptability and effectiveness of the Conv-DMSA model for imputing missing data. The model outperforms baseline methods by significantly reducing the Root Mean Square Error (RMSE) metric, showcasing its superior performance. Specifically, Conv-DMSA has been found to reduce RMSE by 37.2% to 63.87% compared to other models, indicating its enhanced accuracy and efficiency in handling missing data in multivariate time series. Originality/value: The Conv-DMSA model introduces a unique combination of convolutional networks and self-attention mechanisms to the field of missing data imputation. Its innovative use of a diagonal mask within the self-attention block allows for a more nuanced understanding of the data’s temporal and relational aspects. This novel approach not only addresses the existing shortcomings of conventional imputation methods but also sets a new standard for handling missing data in complex, multivariate time series data sets. The model’s superior performance and its capacity to adapt to varying levels of missing data make it a significant contribution to the field.
Toward fast network intrusion detection for web services: partial-flow feature extraction and dataset construction
Purpose: Timely intrusion detection in extensive traffic remains a pressing and complex challenge, including for Web services. Current research emphasizes improving detection accuracy through machine learning, with scant attention paid to the dataset’s impact on the capability for fast detection. Many datasets rely on flow-level features, requiring entire flow completion before determining if it constitutes an attack, reducing efficiency. This paper aims to introduce a new feature extraction method and construct a new security dataset that enhances detection efficiency. Design/methodology/approach: This paper proposes a novel partial-flow feature extraction method that extracts packet-level features efficiently to reduce the high latency of flow-level extraction. The method also integrates statistical and temporal features derived from partial flows to improve accuracy. The method was applied to the original packet capture (PCAP) files utilized in creating the CSE-CIC-IDS 2018 dataset, resulting in the development of the WKLIN-WEB-2023 dataset specifically designed for web intrusion detection. The effectiveness of this method was evaluated by training nine classification models on both the WKLIN-WEB-2023 and CSE-CIC-IDS 2018 datasets. Findings: The experimental results show that models trained on the WKLIN-WEB-2023 dataset consistently outperform those on the CSE-CIC-IDS 2018 dataset across precision, recall, f1-score, and detection latency. This demonstrates the superior effectiveness of the new dataset in enhancing both the efficiency and accuracy of intrusion detection. Originality/value: This study proposes the partial-flow feature extraction method, creating the WKLIN-WEB-2023 dataset. This novel approach significantly enhances detection efficiency while maintaining classification performance, providing a valuable foundation for further research on intrusion detection efficiency.
A text classification method combining in-domain pre-training and prompt learning for the steel e-commerce industry
Purpose: With the development of Web information systems, steel e-commerce platforms have accumulated a large number of quality objection texts. These texts reflect consumer dissatisfaction with the dimensions, appearance and performance of steel products, providing valuable insights for product improvement and consumer decision-making. Currently, mainstream solutions rely on pre-trained models, but their performance on domain-specific data sets and few-shot data sets is not satisfactory. This paper aims to address these challenges by proposing more effective methods for improving model performance on these specialized data sets. Design/methodology/approach: This paper presents a method on the basis of in-domain pre-training, bidirectional encoder representation from Transformers (BERT) and prompt learning. Specifically, a domain-specific unsupervised data set is introduced into the BERT model for in-domain pre-training, enabling the model to better understand specific language patterns in the steel e-commerce industry, enhancing the model’s generalization capability; the incorporation of prompt learning into the BERT model enhances attention to sentence context, improving classification performance on few-shot data sets. Findings: Through experimental evaluation, this method demonstrates superior performance on the quality objection data set, achieving a Macro-F1 score of 93.32%. Additionally, ablation experiments further validate the significant advantages of in-domain pre-training and prompt learning in enhancing model performance. Originality/value: This study clearly demonstrates the value of the new method in improving the classification of quality objection texts for steel products. The findings of this study offer practical insights for product improvement in the steel industry and provide new directions for future research on few-shot learning and domain-specific models, with potential applications in other fields.
Graph-based rank aggregation: a deep-learning approach
Purpose: This study aims to introduce a novel rank aggregation algorithm that leverages graph theory and deep-learning to improve the accuracy and relevance of aggregated rankings in metasearch scenarios, particularly when faced with inconsistent and low-quality rank lists. By strategically selecting a subset of base rankers, the algorithm enhances the quality of the aggregated ranking while using only a subset of base rankers. Design/methodology/approach: The proposed algorithm leverages a graph-based model to represent the interrelationships between base rankers. By applying Spectral clustering, the algorithm identifies a subset of top-performing base rankers based on their retrieval effectiveness. These selected rankers are then integrated into a sequential deep-learning model to estimate relevance labels for query-document pairs. Findings: Empirical evaluation on the MQ2007-agg and MQ2008-agg data sets demonstrates the substantial performance gains achieved by the proposed algorithm compared to baseline methods, with an average improvement of 8.7% in MAP and 11.9% in NDCG1. The algorithm’s effectiveness can be attributed to its ability to effectively integrate diverse perspectives from base rankers and capture complex relationships within the data. Originality/value: This research presents a novel approach to rank aggregation that integrates graph theory and deep-learning. The author proposes a graph-based model to select the most effective subset for metasearch applications by constructing a similarity graph of base rankers. This innovative method addresses the challenges posed by inconsistent and low-quality rank lists, offering a unique solution to the problem.
FS2M: fuzzy smart IoT device pairing protocol via speak to microphone
Purpose: This paper aims to provide a secure and efficient pairing protocol for two devices. Due to the large amount of data involving sensitive information transmitted in Internet of Things (IoT) devices, generating a secure shared key between smart devices for secure data sharing becomes essential. However, existing smart devices pairing schemes require longer pairing time and are difficult to resist attacks caused by context, as the secure channel is established based on restricted entropy from physical context. Design/methodology/approach: This paper proposes a fuzzy smart IoT device pairing protocol via speak to microphone, FS2M. In FS2M, the device pairing is realized from the speaking audio of humans in the environment around the devices, which is easily implemented in the vast majority of Internet products. Specifically, to protect the privacy of secret keys and improve efficiency, this paper presents a single-round pairing protocol by adopting a recently published asymmetric fuzzy encapsulation mechanism (AFEM), which allows devices with similar environmental fingerprints to successfully negotiate the shared key. To instantiate AFEM, this paper presents a construction algorithm, the AFEM-ECC, based on elliptic curve cryptography. Findings: This paper analyzes the security of the FS2M and its pairing efficiency with extensive experiments. The results show that the proposed protocol can achieve a secure device pairing between two IoT devices with high efficiency. Originality/value: In FS2M, a novel cryptographic primitive (i.e., AFEM-ECC) are designed for IoT device pairing by using a new context-environment (i.e., human voice) . The experimental results show that FS2M has a good performance in both communication cost (i.e., 130 KB) and running time (i.e., 10 S).
LLMSARec: large language model with semantic alignment for Web service recommendation
Purpose: With the continuous increase in Web services, efficient identification of Web services that meet developers’ needs and understanding their relationships remains a challenge. Previous research has improved recommendation effectiveness by using correlations between Web services through graph neural networks (GNNs), while it has not fully leveraged service descriptions, limiting the depth and diversity of learning. To this end, a Web services recommendation method called LLMSARec, based on Large Language Model and semantic alignment, is proposed. This study aims to extract potential semantic information from services and learn deeper relationships between services. Design/methodology/approach: This method consists of two core modules: profile generation and maximizing mutual information. The profile generation module uses LLM to analyze the descriptions of services, infer and construct service profiles. Concurrently, it uses LLM as text encoders to encode inferred service profiles for enhanced service representation learning. The maximizing mutual information model aims to align the semantic features of the services text inferred by LLM with structural semantic features of the services captured by GNNs, thus achieving a more comprehensive representation of services. The aligned representation serves as an input for the model to identify services with superior matching accuracy, thereby enhancing the service recommendation capability. Findings: Experimental comparisons and analyses were conducted on the Programmable Web platform data set, and the results demonstrated that the effectiveness of Web service recommendations can be significantly improved by using LLMSARec. Originality/value: In this study, the authors propose a Web service recommendation approach based on Large Language Model and semantic alignment. By extracting latent semantic information from services and effectively aligning semantic features with structural features, new representations can be generated to significantly enhance recommendation accuracy.
Point-of-interest recommendation based on the spatial-temporal graph
Purpose: Point-of-interest (POI) recommendation techniques play a crucial role in mitigating information overload and delivering tailored services. To address limitations in conventional POI recommendation systems, constrained by sparse user-POI interactions and incomplete consideration of temporal dynamics, POI recommendation based on the spatial-temporal graph (STG-POI) is proposed. Design/methodology/approach: Spatial-temporal sequence graphs from geographical locations and user interaction history data are constructed, which are used to mine spatial-temporal sequence information. Using the data filtered by the band-pass filter, graph neural networks with distance-awareness and sequence-awareness are applied to capture high-order spatial-temporal connections within diverse graph topologies. The model leverages contrastive learning for self-supervised disentanglement of graph representations, providing self-supervised signals for sequential and geographical intent perception, thereby achieving more precise POI personalization. Findings: Compared to the baseline model GSTN, experiments on the Foursquare and Gowalla data sets reveal that STG-POI improves testing AUC by 2.0%, 2.1%, 2.0% and decreases logloss by 1.9%, 3.3%, 0.3%, respectively. These results indicate the model’s effectiveness in capturing spatial-temporal information, surpassing mainstream POI recommendation baseline models. Originality/value: This approach constructs a dual graph from user interaction data, harnessing sequential and geographical information as self-supervised signals. It yields decoupled representations of these influences, offering a comprehensive insight into user behaviors and preferences within location-based social networks, thus enhancing recommendation accuracy and interpretability. This approach addresses the challenge in graph convolutional network where only rough and smooth features are conducive to recommendation by using band-pass filters to significantly reduce computational complexity, thereby enhancing recommendation speed by filtering out noise data that does not contribute to recommendation performance. Experimental results indicate that this model surpasses current mainstream approaches in POI recommendation tasks, effectively integrating both geographical and temporal features.
Visual software defect prediction method based on improved recurrent criss-cross residual network
Purpose: This study aims to solve the problems of large training sample size, low data sample quality, low efficiency of the currently used classical model, high computational complexity of the existing concern mechanism, and high graphics processing unit (GPU) occupancy in the current visualization software defect prediction, proposing a method for software defect prediction termed recurrent criss-cross attention for weighted activation functions of recurrent SE-ResNet (RCCA-WRSR). First, following code visualization, the activation functions of the SE-ResNet model are replaced with a weighted combination of Relu and Elu to enhance model convergence. Additionally, an SE module is added before it to filter feature information, eliminating low-weight features to generate an improved residual network model, WRSR. To focus more on contextual information and establish connections between a pixel and those not in the same cross-path, the visualized red as integer, green as integer, blue as integer images are inputted into a model incorporating a fused RCCA module for defect prediction. Design/methodology/approach: Software defect prediction based on code visualization is a new software defect prediction technology, which mainly realizes the defect prediction of code by visualizing code as image, and then applying attention mechanism to extract the features of image. However, the challenges of current visualization software defect prediction mainly include the large training sample size and low sample quality of the data, and the classical models used today are not efficient, and the existing attention mechanisms have high computational complexity and high GPU occupancy. Findings: Experimental evaluation using ten open-source Java data sets from PROMISE and five existing methods demonstrates that the proposed approach achieves an F-measure value of 0.637 in predicting 16 cross-version projects, representing a 6.1% improvement. Originality/value: RCCA-WRSR is a new visual software defect prediction based on recurrent criss-cross attention and improved residual network. This method effectively enhances the performance of software defect prediction.
SEN-CTD: semantic enhancement network with content-title discrepancy for fake news detection
Purpose: The discrepancy between the content of an article and its title is a key characteristic of fake news. Current methods for detecting fake news often ignore the significant difference in length between the content and its title. In addition, relying solely on textual discrepancies between the title and content to distinguish between real and fake news has proven ineffective. The purpose of this paper is to develop a new approach called semantic enhancement network with content–title discrepancy (SEN–CTD), which enhances the accuracy of fake news detection. Design/methodology/approach: The SEN–CTD framework is composed of two primary modules: the SEN and the content–title comparison network (CTCN). The SEN is designed to enrich the representation of news titles by integrating external information and position information to capture the context. Meanwhile, the CTCN focuses on assessing the consistency between the content of news articles and their corresponding titles examining both emotional tones and semantic attributes. Findings: The SEN–CTD model performs well on the GossipCop, PolitiFact and RealNews data sets, achieving accuracies of 80.28%, 86.88% and 84.96%, respectively. These results highlight its effectiveness in accurately detecting fake news across different types of content. Originality/value: The SEN is specifically designed to improve the representation of extremely short texts, enhancing the depth and accuracy of analyses for brief content. The CTCN is tailored to examine the consistency between news titles and their corresponding content, ensuring a thorough comparative evaluation of both emotional and semantic discrepancies.
Leveraging recommendations using a multiplex graph database
Purpose: By applying targeted graph algorithms, the method used by the authors enables effective prediction of user interactions and thus fulfils the complex requirements of modern recommender systems. This study sets a new benchmark for multidimensional recommendation strategies and offers a path towards more advanced and user-centric models. Design/methodology/approach: To improve multidimensional data recommendation systems, multiplex graph structures are useful to capture various types of user interactions. This paper presents a novel framework that uses a graph database to compute and manipulate multiplex graphs. The approach enables flexible dimension management and increases expressive power through a specialised algebra designed for multiplex graph manipulation. Findings: The authors compare the multiplex graph approach with traditional matrix methods, in particular random walk with restart, and show that the method not only provides deeper insights into user preferences by integrating scores from different layers of the multiplex graph, but also outperforming matrix-based approaches in most configurations. The results highlight the potential of multiplex graphs for developing sophisticated and customised recommender systems that significantly improve both performance and explainability. Originality/value: The study provides a formal specification of a multiplex graph construction based on interaction and content-based information; and the study also developed an algebra dedicated to multiplex graphs, enabling robust and precise graph manipulations necessary for effective recommendation queries. The authors implement these algebraic operations within the Neo4j graph database system with a thorough analysis and experimentation with three different data sets, benchmarked against traditional matrix-based methods.