Pub Date : 2025-09-01DOI: 10.1016/j.jiixd.2024.09.001
Tinshu Sasi , Arash Habibi Lashkari , Rongxing Lu , Pulei Xiong , Shahrear Iqbal
In the last ten years, the IoT has played a crucial role in society's digital transformation. However, because of the wide range of devices it encompasses, it is also facing increased security vulnerabilities. This research presents a novel mechanism called the self-attention-based 1D-CNN-LSTM, which uses convolutional neural networks (CNNs) combined with a long short-term memory (LSTM) model enhanced with a self-attention mechanism for detecting IoT attacks. The proposed mechanism achieves high accuracy and efficiently differentiates malicious and benign network traffic. By employing Shapley additive explanations (SHAP), we identified important predictive features from the preprocessed data, which were retrieved using CICFlowMeter. This has strengthened the dependability of the model. In addition, we enhanced the model by training it on a smaller collection of features, resulting in shorter training time while preserving accuracy. We have also generated nine augmented IoT tabular datasets named CIC-BCCC-NRC_TabularIoTAttack-2024 from accessible IoT datasets to evaluate the model's robustness and showcase its efficacy in IoT security.
{"title":"An efficient self attention-based 1D-CNN-LSTM network for IoT attack detection and identification using network traffic","authors":"Tinshu Sasi , Arash Habibi Lashkari , Rongxing Lu , Pulei Xiong , Shahrear Iqbal","doi":"10.1016/j.jiixd.2024.09.001","DOIUrl":"10.1016/j.jiixd.2024.09.001","url":null,"abstract":"<div><div>In the last ten years, the IoT has played a crucial role in society's digital transformation. However, because of the wide range of devices it encompasses, it is also facing increased security vulnerabilities. This research presents a novel mechanism called the self-attention-based 1D-CNN-LSTM, which uses convolutional neural networks (CNNs) combined with a long short-term memory (LSTM) model enhanced with a self-attention mechanism for detecting IoT attacks. The proposed mechanism achieves high accuracy and efficiently differentiates malicious and benign network traffic. By employing Shapley additive explanations (SHAP), we identified important predictive features from the preprocessed data, which were retrieved using CICFlowMeter. This has strengthened the dependability of the model. In addition, we enhanced the model by training it on a smaller collection of features, resulting in shorter training time while preserving accuracy. We have also generated nine augmented IoT tabular datasets named CIC-BCCC-NRC_TabularIoTAttack-2024 from accessible IoT datasets to evaluate the model's robustness and showcase its efficacy in IoT security.</div></div>","PeriodicalId":100790,"journal":{"name":"Journal of Information and Intelligence","volume":"3 5","pages":"Pages 375-400"},"PeriodicalIF":0.0,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145371322","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In recent years, the right to deletion of individual has been recognized by many privacy protection laws and regulations. It stipulates that the data controller receiving individual data deletion request shall not only erase the required data, but also take reasonable steps to inform other data controllers to delete the same data. Prior to irrecoverable data erasure, it is of paramount importance to design a distribution and acknowledgement process of deletion notifications across involved data controllers. The design of such a mechanism is faced with the following challenges: 1) completeness: Ensuring that all relevant data controllers, who possess the data slated for erasure, are duly informed; 2) robustness: Immune from malicious attacks when deletion notifications traverse through untrusted networks; 3) lightweight: Reduce the right to deletion compliance cost and accommodate more deletion requests for data controllers. To this end, this article proposes DelRightGuard, which is the first attempt to tackle with the aforementioned challenges. DelRightGuard is built on a cross-plane cooperation architecture between regulatory and service planes. Within regulatory plane, DelRightGuard proposes a cuckoo filters based on data circulation recording algorithm to efficiently ensure the completeness of deletion notifications. Within service plane, DelRightGuard devises a secure yet lightweight deletion notification distribution protocol that runs on a network function hosted by each data controller. This protocol employs HMAC based hop-by-hop forward traversal verification, recursive backward acknowledgement and probabilistic sampling verification, so that it ensure the robustness and lightweight of deletion notification distribution process. We implement a prototype for DelRightGuard. The experimental result confirms that it is practical with acceptable performance.
{"title":"DelRightGuard: A secure yet lightweight data deletion notification distribution protocol for safeguarding right to deletion","authors":"Qipeng Song, Ruiyun Wang, Yue Li, Yiheng Yan, Xingyue Zhu, Hui Li","doi":"10.1016/j.jiixd.2024.11.001","DOIUrl":"10.1016/j.jiixd.2024.11.001","url":null,"abstract":"<div><div>In recent years, the right to deletion of individual has been recognized by many privacy protection laws and regulations. It stipulates that the data controller receiving individual data deletion request shall not only erase the required data, but also take reasonable steps to inform other data controllers to delete the same data. Prior to irrecoverable data erasure, it is of paramount importance to design a distribution and acknowledgement process of deletion notifications across involved data controllers. The design of such a mechanism is faced with the following challenges: 1) completeness: Ensuring that all relevant data controllers, who possess the data slated for erasure, are duly informed; 2) robustness: Immune from malicious attacks when deletion notifications traverse through untrusted networks; 3) lightweight: Reduce the right to deletion compliance cost and accommodate more deletion requests for data controllers. To this end, this article proposes DelRightGuard, which is the first attempt to tackle with the aforementioned challenges. DelRightGuard is built on a cross-plane cooperation architecture between regulatory and service planes. Within regulatory plane, DelRightGuard proposes a cuckoo filters based on data circulation recording algorithm to efficiently ensure the completeness of deletion notifications. Within service plane, DelRightGuard devises a secure yet lightweight deletion notification distribution protocol that runs on a network function hosted by each data controller. This protocol employs HMAC based hop-by-hop forward traversal verification, recursive backward acknowledgement and probabilistic sampling verification, so that it ensure the robustness and lightweight of deletion notification distribution process. We implement a prototype for DelRightGuard. The experimental result confirms that it is practical with acceptable performance.</div></div>","PeriodicalId":100790,"journal":{"name":"Journal of Information and Intelligence","volume":"3 4","pages":"Pages 303-325"},"PeriodicalIF":0.0,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144895163","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-07-01DOI: 10.1016/j.jiixd.2025.04.003
Lang Qin , Yi Liu , Caihong Mu
In the era of information technology, recommendation systems play a crucial role in information filtering and user preference identification. Notably, the auxiliary information provided by online social platforms offers significant support for enhancing the performance of recommendation systems. Based on the hypothesis that socially connected users share similar preferences, integrating social relationships as supplementary information into recommendation algorithms can significantly enhance recommendation accuracy while effectively mitigating the cold-start problem. However, existing social recommendation systems primarily rely on explicit social relationships as auxiliary information, often overlooking the value of potential social connections. Research indicates that users with potential social links may also possess valuable preference information. We believe that mining potential social relationships can provide valuable auxiliary information, thereby enhancing the performance of recommendation systems. To address this issue, we propose a social recommendation model based on social semantic mining and denoising (SSMD). Specifically, we propose an encoder-decoder architecture to learn explicit social user representations and mine potential social relationships. Considering the potential noise in these implicit connections, we design a denoising module that utilizes user preference information to filter unreliable social links. Furthermore, we implement cross-view information alignment between the potential social graph and interaction graph through an auxiliary loss function. Extensive experiments conducted on multiple public datasets demonstrate that our SSMD method outperforms various baseline approaches with significant improvements.
{"title":"A social recommendation model based on social semantic mining and denoising","authors":"Lang Qin , Yi Liu , Caihong Mu","doi":"10.1016/j.jiixd.2025.04.003","DOIUrl":"10.1016/j.jiixd.2025.04.003","url":null,"abstract":"<div><div>In the era of information technology, recommendation systems play a crucial role in information filtering and user preference identification. Notably, the auxiliary information provided by online social platforms offers significant support for enhancing the performance of recommendation systems. Based on the hypothesis that socially connected users share similar preferences, integrating social relationships as supplementary information into recommendation algorithms can significantly enhance recommendation accuracy while effectively mitigating the cold-start problem. However, existing social recommendation systems primarily rely on explicit social relationships as auxiliary information, often overlooking the value of potential social connections. Research indicates that users with potential social links may also possess valuable preference information. We believe that mining potential social relationships can provide valuable auxiliary information, thereby enhancing the performance of recommendation systems. To address this issue, we propose a social recommendation model based on social semantic mining and denoising (SSMD). Specifically, we propose an encoder-decoder architecture to learn explicit social user representations and mine potential social relationships. Considering the potential noise in these implicit connections, we design a denoising module that utilizes user preference information to filter unreliable social links. Furthermore, we implement cross-view information alignment between the potential social graph and interaction graph through an auxiliary loss function. Extensive experiments conducted on multiple public datasets demonstrate that our SSMD method outperforms various baseline approaches with significant improvements.</div></div>","PeriodicalId":100790,"journal":{"name":"Journal of Information and Intelligence","volume":"3 4","pages":"Pages 361-374"},"PeriodicalIF":0.0,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144895166","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-07-01DOI: 10.1016/j.jiixd.2025.03.005
Yangyang Cai, Sulan Li, Yongliang Wei, Yunru Zhu
Aiming at the consensus of relative position considering obstacle avoidance for fractional-order multi-agent system, a novel distributed control algorithm is proposed in this paper. Firstly, a synthetic error of each agent under the influence of obstacles is introduced. The consensus protocols are designed based on this error according to sliding mode theory for the order increasing and decreasing, respectively. Then, the Lyapunov function is used to prove the stable convergence of the protocols. Finally, the simulation results show that the protocols can not only prevent the agents from colliding with obstacles, but also enable the agents to quickly recover the expected formation and achieve consensus of the relative position.
{"title":"Time-varying formation control with obstacle avoidance for fractional-order multi-agent systems","authors":"Yangyang Cai, Sulan Li, Yongliang Wei, Yunru Zhu","doi":"10.1016/j.jiixd.2025.03.005","DOIUrl":"10.1016/j.jiixd.2025.03.005","url":null,"abstract":"<div><div>Aiming at the consensus of relative position considering obstacle avoidance for fractional-order multi-agent system, a novel distributed control algorithm is proposed in this paper. Firstly, a synthetic error of each agent under the influence of obstacles is introduced. The consensus protocols are designed based on this error according to sliding mode theory for the order increasing and decreasing, respectively. Then, the Lyapunov function is used to prove the stable convergence of the protocols. Finally, the simulation results show that the protocols can not only prevent the agents from colliding with obstacles, but also enable the agents to quickly recover the expected formation and achieve consensus of the relative position.</div></div>","PeriodicalId":100790,"journal":{"name":"Journal of Information and Intelligence","volume":"3 4","pages":"Pages 289-302"},"PeriodicalIF":0.0,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144895148","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-07-01DOI: 10.1016/j.jiixd.2025.03.002
Weidong Wan, Yi Liu, Hailin Zhang
Symbiotic radio (SR) is a technology that facilitates mutually beneficial sharing of spectrum and energy between primary and secondary systems. In SR networks, utilizing active reconfigurable intelligent surface (RIS) as the secondary transmitter (STx) enhances this mutual benefit compared to passive RIS. This paper addresses the interference management challenges that inevitably arise from employing active RIS. We consider a common SR network consisting of three types of users: SR users, non-SR users, and eavesdroppers. Additionally, each SR user has their own unique cellular services. We propose minimizing the total power consumption while satisfying a sufficiently large signal-to-interference-plus-noise ratio (SINR) for SR users, a small enough SINR for eavesdroppers, and a small enough interference temperature for non-SR users. The alternative optimization (AO) method is used for decoupling multi-variables. The non-convex constraints are relaxed as convex ones through first-order Taylor approximation, and the bounded channel state information (CSI) error model is handled using the S-procedure. Simulations validate the superiority of the proposed algorithm and demonstrate that the total power consumption is minimized while meeting performance thresholds. Additionally, the results offer valuable insights for SR network deployment.
{"title":"Interference management for active RIS-aided symbiotic radio networks","authors":"Weidong Wan, Yi Liu, Hailin Zhang","doi":"10.1016/j.jiixd.2025.03.002","DOIUrl":"10.1016/j.jiixd.2025.03.002","url":null,"abstract":"<div><div>Symbiotic radio (SR) is a technology that facilitates mutually beneficial sharing of spectrum and energy between primary and secondary systems. In SR networks, utilizing active reconfigurable intelligent surface (RIS) as the secondary transmitter (STx) enhances this mutual benefit compared to passive RIS. This paper addresses the interference management challenges that inevitably arise from employing active RIS. We consider a common SR network consisting of three types of users: SR users, non-SR users, and eavesdroppers. Additionally, each SR user has their own unique cellular services. We propose minimizing the total power consumption while satisfying a sufficiently large signal-to-interference-plus-noise ratio (SINR) for SR users, a small enough SINR for eavesdroppers, and a small enough interference temperature for non-SR users. The alternative optimization (AO) method is used for decoupling multi-variables. The non-convex constraints are relaxed as convex ones through first-order Taylor approximation, and the bounded channel state information (CSI) error model is handled using the S-procedure. Simulations validate the superiority of the proposed algorithm and demonstrate that the total power consumption is minimized while meeting performance thresholds. Additionally, the results offer valuable insights for SR network deployment.</div></div>","PeriodicalId":100790,"journal":{"name":"Journal of Information and Intelligence","volume":"3 4","pages":"Pages 326-344"},"PeriodicalIF":0.0,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144895164","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-07-01DOI: 10.1016/j.jiixd.2025.03.004
Min Ouyang , Tong Wang , Pei Xiao , Jiyi Wu , Shan Gao , Liwei Chen
Interference range plays a critical role in wireless network performance, significantly impacting both link reliability and resource utilization. This paper studies the interference range associated with the spatial outage capacity (SOC), which is the maximum density of reliable links of bipolar networks. We establish a recursive equation based on the transmitter's active probability, establishing a link between the interference range and the SOC. The analytical results are then verified through numerical and network simulations. The experimental results indicate that the interference range may improve the SOC of Poisson bipolar networks while deteriorating the SOC of Poisson cellular networks and random distance bipolar networks.
{"title":"The interference range of the spatial outage capacity in bipolar wireless networks","authors":"Min Ouyang , Tong Wang , Pei Xiao , Jiyi Wu , Shan Gao , Liwei Chen","doi":"10.1016/j.jiixd.2025.03.004","DOIUrl":"10.1016/j.jiixd.2025.03.004","url":null,"abstract":"<div><div>Interference range plays a critical role in wireless network performance, significantly impacting both link reliability and resource utilization. This paper studies the interference range associated with the spatial outage capacity (SOC), which is the maximum density of reliable links of bipolar networks. We establish a recursive equation based on the transmitter's active probability, establishing a link between the interference range and the SOC. The analytical results are then verified through numerical and network simulations. The experimental results indicate that the interference range may improve the SOC of Poisson bipolar networks while deteriorating the SOC of Poisson cellular networks and random distance bipolar networks.</div></div>","PeriodicalId":100790,"journal":{"name":"Journal of Information and Intelligence","volume":"3 4","pages":"Pages 345-360"},"PeriodicalIF":0.0,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144895165","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-05-01DOI: 10.1016/j.jiixd.2025.03.001
Wengang Li, Deli Zhou, Qiong Ye
Orthogonal time-frequency space (OTFS) modulation can effectively counter ICI in high-speed mobile scenarios, fully enhance the spectral efficiency of communication systems in high Doppler expansion scenarios, and improve the quality of communication systems. Channel estimation performance serves as a critical evaluation parameter within the OTFS modulation system. In this paper, we propose a multi-scale attention residual neural structure for improved channel estimation of OTFS waveforms in different satellite-ground scenario. Firstly, a multi-scale channel feature extraction module is designed, which applies multi-dimensional feature extraction to the channel matrix, thereby bolstering the capability to capture features at diverse scales. Subsequently, a self-attention mechanism is incorporated to concentrate on subtle yet significant features. The extracted features are then integrated and exploited through a residual convolutional architecture to derive an estimation of the channel matrix. Simulations are conducted using the satellite-ground mobile channel model outlined in 3GPP TR 38.811, with the NTN-TDL-C and NTN-TDL-B channel models representing line of sight (LoS) and non-line of sight (NLoS) conditions, respectively. Results demonstrate that the attention-based approach presented surpasses alternative neural network methodologies in terms of mean squared error (MSE), bit error rate (BER), and complexity, and meets the demands of OTFS channel estimation in satellite-ground scenario.
{"title":"Channel computation based on multi-scale attention residual network","authors":"Wengang Li, Deli Zhou, Qiong Ye","doi":"10.1016/j.jiixd.2025.03.001","DOIUrl":"10.1016/j.jiixd.2025.03.001","url":null,"abstract":"<div><div>Orthogonal time-frequency space (OTFS) modulation can effectively counter ICI in high-speed mobile scenarios, fully enhance the spectral efficiency of communication systems in high Doppler expansion scenarios, and improve the quality of communication systems. Channel estimation performance serves as a critical evaluation parameter within the OTFS modulation system. In this paper, we propose a multi-scale attention residual neural structure for improved channel estimation of OTFS waveforms in different satellite-ground scenario. Firstly, a multi-scale channel feature extraction module is designed, which applies multi-dimensional feature extraction to the channel matrix, thereby bolstering the capability to capture features at diverse scales. Subsequently, a self-attention mechanism is incorporated to concentrate on subtle yet significant features. The extracted features are then integrated and exploited through a residual convolutional architecture to derive an estimation of the channel matrix. Simulations are conducted using the satellite-ground mobile channel model outlined in 3GPP TR 38.811, with the NTN-TDL-C and NTN-TDL-B channel models representing line of sight (LoS) and non-line of sight (NLoS) conditions, respectively. Results demonstrate that the attention-based approach presented surpasses alternative neural network methodologies in terms of mean squared error (MSE), bit error rate (BER), and complexity, and meets the demands of OTFS channel estimation in satellite-ground scenario.</div></div>","PeriodicalId":100790,"journal":{"name":"Journal of Information and Intelligence","volume":"3 3","pages":"Pages 275-287"},"PeriodicalIF":0.0,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144490380","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-05-01DOI: 10.1016/j.jiixd.2025.02.006
Qiang Yu , Liang Bao , Peng Nie , Lei Zuo
Knowledge graph completion aims to enhance the completeness of knowledge graphs by predicting missing links. Link prediction is a common approach for this task, but existing methods, particularly those based on similarity computation, are often computationally expensive, especially for large models. To address this, we propose a novel method, positionally restricted masked knowledge graph completion (PR-MKGC), which reduces inference time by leveraging masked prediction and relying solely on structural information from the knowledge graph, without using textual data. We introduce a multi-head mutual attention mechanism that aggregates neighbor information more effectively, improving the model's ability to predict missing links. Experimental results demonstrate that PR-MKGC outperforms existing models in terms of both predictive performance and inference time on the FB15K-237 dataset.
{"title":"Positionally restricted masked knowledge graph completion via multi-head mutual attention","authors":"Qiang Yu , Liang Bao , Peng Nie , Lei Zuo","doi":"10.1016/j.jiixd.2025.02.006","DOIUrl":"10.1016/j.jiixd.2025.02.006","url":null,"abstract":"<div><div>Knowledge graph completion aims to enhance the completeness of knowledge graphs by predicting missing links. Link prediction is a common approach for this task, but existing methods, particularly those based on similarity computation, are often computationally expensive, especially for large models. To address this, we propose a novel method, positionally restricted masked knowledge graph completion (PR-MKGC), which reduces inference time by leveraging masked prediction and relying solely on structural information from the knowledge graph, without using textual data. We introduce a multi-head mutual attention mechanism that aggregates neighbor information more effectively, improving the model's ability to predict missing links. Experimental results demonstrate that PR-MKGC outperforms existing models in terms of both predictive performance and inference time on the FB15K-237 dataset.</div></div>","PeriodicalId":100790,"journal":{"name":"Journal of Information and Intelligence","volume":"3 3","pages":"Pages 210-222"},"PeriodicalIF":0.0,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144490376","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-05-01DOI: 10.1016/j.jiixd.2025.04.001
Yunfei Zhao , Jie Guo , Longyu Wen , Letian Wang
In recent years, there has been a burgeoning interest in multimodal recommender systems, which integrate various data types to achieve more personalized recommendations. Despite this, the effective incorporation of user preferences for multimodal data and the exploration of inherent semantic relationships between modalities still need to be explored. Prior research typically utilizes multimodal data to construct item graphs, often overlooking the nuanced details within the data. As a result, these studies fail to thoroughly examine the semantic relationships between items and user behavioral patterns. Our proposed approach, METRIC, addresses this gap by delving deeper into multimodal information. METRIC consists of two primary modules: the multiple preference modelling (MPM) module and the item semantic enhancement (ISE) module. The ISE module performs relational mining across multiple attributes, leveraging the semantic structural relationships inherent in items. In contrast, the MPM module enables users to articulate their preferences across different modalities and facilitates adaptive fusion through an attention mechanism. This approach not only improves precision in capturing user preferences and interests but also minimizes interference from varying modalities. Our extensive experiments on three benchmark datasets substantiate METRIC's superiority and the efficacy of its core components.
{"title":"METRIC: Multiple preferences learning with refined item attributes for multimodal recommendation","authors":"Yunfei Zhao , Jie Guo , Longyu Wen , Letian Wang","doi":"10.1016/j.jiixd.2025.04.001","DOIUrl":"10.1016/j.jiixd.2025.04.001","url":null,"abstract":"<div><div>In recent years, there has been a burgeoning interest in multimodal recommender systems, which integrate various data types to achieve more personalized recommendations. Despite this, the effective incorporation of user preferences for multimodal data and the exploration of inherent semantic relationships between modalities still need to be explored. Prior research typically utilizes multimodal data to construct item graphs, often overlooking the nuanced details within the data. As a result, these studies fail to thoroughly examine the semantic relationships between items and user behavioral patterns. Our proposed approach, METRIC, addresses this gap by delving deeper into multimodal information. METRIC consists of two primary modules: the multiple preference modelling (MPM) module and the item semantic enhancement (ISE) module. The ISE module performs relational mining across multiple attributes, leveraging the semantic structural relationships inherent in items. In contrast, the MPM module enables users to articulate their preferences across different modalities and facilitates adaptive fusion through an attention mechanism. This approach not only improves precision in capturing user preferences and interests but also minimizes interference from varying modalities. Our extensive experiments on three benchmark datasets substantiate METRIC's superiority and the efficacy of its core components.</div></div>","PeriodicalId":100790,"journal":{"name":"Journal of Information and Intelligence","volume":"3 3","pages":"Pages 242-256"},"PeriodicalIF":0.0,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144490378","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-05-01DOI: 10.1016/j.jiixd.2025.02.002
Xiang-Gen Xia
In this paper, we rethink delay Doppler channels (also called doubly selective channels). We prove that no modulation schemes (including the current active VOFDM/OTFS) can compensate a non-trivial Doppler spread well. We then discuss some of the existing methods to deal with time-varying channels, in particular time-frequency (TF) coding in an OFDM system. TF coding is equivalent to space-time coding in the math part. We also summarize state of the art on space-time coding that was an active research topic over a decade ago.
{"title":"Rethink delay Doppler channels and time-frequency coding","authors":"Xiang-Gen Xia","doi":"10.1016/j.jiixd.2025.02.002","DOIUrl":"10.1016/j.jiixd.2025.02.002","url":null,"abstract":"<div><div>In this paper, we rethink delay Doppler channels (also called doubly selective channels). We prove that no modulation schemes (including the current active VOFDM/OTFS) can compensate a non-trivial Doppler spread well. We then discuss some of the existing methods to deal with time-varying channels, in particular time-frequency (TF) coding in an OFDM system. TF coding is equivalent to space-time coding in the math part. We also summarize state of the art on space-time coding that was an active research topic over a decade ago.</div></div>","PeriodicalId":100790,"journal":{"name":"Journal of Information and Intelligence","volume":"3 3","pages":"Pages 189-193"},"PeriodicalIF":0.0,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144490374","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号