Knowledge-Based Systems最新文献

Progressive de-preference task-specific processing for generalizable person re-identification

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-29 DOI: 10.1016/j.knosys.2024.112779

Haishun Du, Jieru Li, Linbing Cao, Xinxin Hao

{"title":"Progressive de-preference task-specific processing for generalizable person re-identification","authors":"Haishun Du, Jieru Li, Linbing Cao, Xinxin Hao","doi":"10.1016/j.knosys.2024.112779","DOIUrl":"10.1016/j.knosys.2024.112779","url":null,"abstract":"<div><div>Recently, domain generalization (DG) person re-identification (ReID) has attracted attention. Existing DG person ReID methods train on mixed datasets containing all source domains. However, these mixed datasets have huge inter-domain differences because of varying data distributions across different source domains. Such differences hinder models from learning domain-invariant representations, affecting generalization on unseen domains. To address this issue, we propose a progressive de-preference task-specific processing network (PDTP-Net) for DG person ReID. Initially, we design a progressive de-preference domain segmentation strategy to mitigate inter-domain differences by dividing multiple source domains into different phases, each comprising several training tasks. We then design a global and task-specific processing module that enhances extraction of domain-invariant features by integrating statistical information from other source domains. Finally, we design a multi-granularity attention module and a group-aware batch normalization strategy to ensure the features are more discriminative and better suited for person ReID tasks. The proposed model is validated using three DG person ReID experimental protocols: Protocol-1, Protocol-2, and leave-one-out experiments. On Protocol-1, the model improves mean average precision (mAP) and Rank-1 accuracy on all datasets by an average of 0.7% and 0.3%, respectively. On Protocol-2, the model improves mAP and Rank-1 accuracy on all datasets by an average of 2.525% and 2.725%, respectively. On the leave-one-out experiments, the model improves mAP and Rank-1 accuracy on all tasks by an average of 0.65% and 0.18%, respectively. The results on several popular datasets suggest that the model achieves state-of-the-art performance in DG person ReID.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":"309 ","pages":"Article 112779"},"PeriodicalIF":7.2,"publicationDate":"2024-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142758824","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

GKA-GPT: Graphical knowledge aggregation for multiturn dialog generation

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-28 DOI: 10.1016/j.knosys.2024.112763

Yuezhou Dong , Ke Qin , Shuang Liang , Ahmad Raza , Guangchun Luo

{"title":"GKA-GPT: Graphical knowledge aggregation for multiturn dialog generation","authors":"Yuezhou Dong , Ke Qin , Shuang Liang , Ahmad Raza , Guangchun Luo","doi":"10.1016/j.knosys.2024.112763","DOIUrl":"10.1016/j.knosys.2024.112763","url":null,"abstract":"<div><div>In human interaction, effective communication relies on shared cognitive processes that facilitate the ability of individuals to comprehend the intended message of their interlocutors. Recent research in multiturn dialog generation seeks to emulate human-like responses by incorporating external knowledge into generative models to enhance language understanding. These models often utilize graphical representations of knowledge and employ graph neural networks (GNNs) to capture dialog semantics. However, sole reliance on external knowledge can fall short as human cognition integrates universal commonsense and personal knowledge, with the latter being derived from individual experiences and frequently disregarded. To remedy this, we propose GKA-GPT, a novel GNN-based approach that merges commonsense and personal knowledge into a comprehensive cognition graph to enhance the relevance and diversity of responses in multiturn dialog scenarios. Furthermore, GKA-GPT introduces a multigrained graphical knowledge aggregation mechanism for effective semantic information processing across various levels. Our experiments demonstrate that GKA-GPT outperforms existing baselines by generating more relevant and informative responses.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":"309 ","pages":"Article 112763"},"PeriodicalIF":7.2,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142748614","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A novel spatio-temporal feature interleaved contrast learning neural network from a robustness perspective

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-28 DOI: 10.1016/j.knosys.2024.112788

Peng Liu , Yaodong Zhu , Yang Yang , Caixia Wang , Mingqiu Li , Haifang Cong , Guangyu Zhao , Han Yang

{"title":"A novel spatio-temporal feature interleaved contrast learning neural network from a robustness perspective","authors":"Peng Liu , Yaodong Zhu , Yang Yang , Caixia Wang , Mingqiu Li , Haifang Cong , Guangyu Zhao , Han Yang","doi":"10.1016/j.knosys.2024.112788","DOIUrl":"10.1016/j.knosys.2024.112788","url":null,"abstract":"<div><div>Accurate traffic forecasting is critical to the effectiveness of intelligent transportation systems (ITS) and the development of smart cities.Achieving this goal requires efficient capture of heterogeneous interactions between spatial and temporal dependencies of traffic nodes.However, the robustness and predictive capacity of modeling systems are frequently compromised by the limitations inherent in fine-grained sensor data collection methodologies.Furthermore, the uneven distribution of data can exacerbate the degradation of the model’s predictive performance.To tackle these challenges, we introduce an innovative neural network that leverages spatio-temporal feature interlace contrast learning for daily traffic flow prediction.Our approach consists of two main parts: First, we propose a spatiotemporal position encoder that aims to provide a more balanced sample of training spatiotemporal data with mixed spatial coding to solve the problem of local heterogeneity in the data.Secondly, we employ a spatiotemporal interlace contrast graph structure generator and a specific structure and direction discriminator to discern various potential spatiotemporal features and categorize samples based on trends and consistency, thereby augmenting the system’s robustness and generalization capabilities. Extensive experiments and case studies across six real datasets demonstrate that our approach markedly enhances the prediction accuracy of the baseline model and introduces novel prediction strategies aimed at boosting the system’s robustness.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":"309 ","pages":"Article 112788"},"PeriodicalIF":7.2,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142748616","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Graph protection under multiple simultaneous attacks: A heuristic approach

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-28 DOI: 10.1016/j.knosys.2024.112791

Marko Djukanović , Stefan Kapunac , Aleksandar Kartelj , Dragan Matić

{"title":"Graph protection under multiple simultaneous attacks: A heuristic approach","authors":"Marko Djukanović , Stefan Kapunac , Aleksandar Kartelj , Dragan Matić","doi":"10.1016/j.knosys.2024.112791","DOIUrl":"10.1016/j.knosys.2024.112791","url":null,"abstract":"<div><div>This work focuses on developing a meta-heuristic approach to protect network nodes from simultaneous attacks, specifically addressing the <span><math><mi>k</mi></math></span>-strong Roman domination problem. The objective is to assign integer weights to the nodes, representing the number of stationed armies, to meet protection constraints while minimizing the total number of armies. A network is protected if it can repel any simultaneous attack on <span><math><mi>k</mi></math></span> nodes. A node is protected if it can defend itself or if a neighboring node provides an army while retaining at least one army for self-defense. This problem formulation can be used in practical scenarios, e.g. developing counter-terrorism strategies or in coping with supply chain disruptions. The problem is difficult as even verifying the feasibility of a single solution generally requires an exponential time. Two exact approaches are proposed in the literature but applicable to small random graphs. For larger graphs, we propose an effective variable neighborhood search, where the feasibility of a solution is verified by introducing the concept of relaxed feasibility. Experiments are conducted with random networks from the literature and two introduced ad-hoc wireless and real-world networks. Extensive experimental evaluations show the robustness of the proposed approach compared to the existing approaches from the literature by significantly outperforming them in all three benchmark sets. Furthermore, we demonstrate the practical application of the proposed variable neighborhood search approach, where its solution is used to position fire stations within the city so that simultaneous fires can be extinguished efficiently while reducing the number of required fire trucks.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":"309 ","pages":"Article 112791"},"PeriodicalIF":7.2,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142758849","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

PSNet: A non-uniform illumination correction method for underwater images based pseudo-siamese network

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-28 DOI: 10.1016/j.knosys.2024.112780

Wenfeng Zhao, Shenghui Rong, Chen Feng, Bo He

{"title":"PSNet: A non-uniform illumination correction method for underwater images based pseudo-siamese network","authors":"Wenfeng Zhao, Shenghui Rong, Chen Feng, Bo He","doi":"10.1016/j.knosys.2024.112780","DOIUrl":"10.1016/j.knosys.2024.112780","url":null,"abstract":"<div><div>Autonomous underwater vehicles (AUVs) based on visual perception play an important role in maritime operations. However, underwater environment often suffers from poor lighting conditions, making the utilization of artificial light sources necessary. This reliance on artificial lighting frequently results in non-uniform illumination. Furthermore, the absorption and scattering effects of water cause further degradation, such as color distortion and blurring of details. To address these challenges, we propose a pseudo-siamese network, named PSNet, designed for underwater optical image enhancement. PSNet separates the non-uniformly illuminated layer from the optimally uniformly illuminated image and utilizes a cascading iteration strategy to enhance the image details. To achieve a better balance prediction quality, we introduce structure loss and residual reconstruction loss as additional guides for model learning. Additionally, we incorporate a color consistency loss to mitigate color distortion. To address the lack of training data, we develop a non-uniform illumination model and generate a dataset that includes both non-uniformly illuminated layers and uniformly illuminated images. Through comprehensive experimental evaluations, PSNet significantly enhances the visual quality of underwater optical images and consistently outperforms state-of-the-art approaches in multiple performance metrics.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":"309 ","pages":"Article 112780"},"PeriodicalIF":7.2,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142757314","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A novel domain-private-suppress meta-recognition network based universal domain generalization for machinery fault diagnosis

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-28 DOI: 10.1016/j.knosys.2024.112775

Mengdi Xu, Yingjie Zhang, Biliang Lu, Zhaolin Liu, Qingshuai Sun

{"title":"A novel domain-private-suppress meta-recognition network based universal domain generalization for machinery fault diagnosis","authors":"Mengdi Xu, Yingjie Zhang, Biliang Lu, Zhaolin Liu, Qingshuai Sun","doi":"10.1016/j.knosys.2024.112775","DOIUrl":"10.1016/j.knosys.2024.112775","url":null,"abstract":"<div><div>Domain generalization aims to generalize knowledge to target domains not seen during the training phase, even in domain gaps. However, in complex industrial settings, the emergence of new fault types is frequent. Concurrently, the rarity of these faults means that the data collected may not fully capture the entire range of potential fault conditions. As a result, it is challenging to ensure that there is an overlap between the label sets of the multi-source domains and the unseen target domains. This problem requires no prior knowledge of label sets, and it requires a model to learn from multi-source domains and perform well on unknown target domains. In this paper, we propose a Domain-Private-Suppress Meta-Recognition Network (DPSMR). It quantifies channel-level transferability to continuously enhance the robustness of channels to domain shifts, thereby promoting the generalization of a common label set. Using an enhanced meta-recognition calibration algorithm to avoid overconfidence in neural network predictions, we ensure the successful recognition of private samples. By employing dual-consistency loss, we reduce channel instability and facilitate learning domain-invariant features. Experimental results on two multi-domain datasets demonstrate that DPSMR outperforms the state-of-the-art methods.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":"309 ","pages":"Article 112775"},"PeriodicalIF":7.2,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142757315","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Enhancing information fusion and feature selection efficiency via the PROMETHEE method for multi-source dynamic decision data sets

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-28 DOI: 10.1016/j.knosys.2024.112781

Weihua Xu, Yigao Li

{"title":"Enhancing information fusion and feature selection efficiency via the PROMETHEE method for multi-source dynamic decision data sets","authors":"Weihua Xu, Yigao Li","doi":"10.1016/j.knosys.2024.112781","DOIUrl":"10.1016/j.knosys.2024.112781","url":null,"abstract":"<div><div>With the surge in big data, the complexity of synthesizing information from multiple sources has become a critical challenge for feature selection methodologies. Feature selection is the process of reducing the number of attributes in data. Traditional single-source centric approaches are inefficient, requiring extensive preprocessing for multi-source data consolidation prior to feature selection. At the same time, an information fusion method is needed to transform the multi-source information system with selected features into a single-source information system. This paper introduces a novel multi-source information fusion and feature selection approach that seamlessly integrates the Preference Ranking Organization Method for Enrichment Evaluations (PROMETHEE) with a dynamic adaptation mechanism. This method is adept at addressing the complexities introduced by the evolving nature of feature and information source dimensions. The Attribute Evaluation Matrix (AEM) and the Attribute Preference Degree Matrix (APDM) are proposed to systematically assess and rank the significance of attributes within a static decision-making framework. Following this, an information fusion method using the source center is proposed. The dynamic feature selection and information fusion methods are proposed to deal with the condition when number of attributes and samples change. Extensive experimental validation confirms that this method not only reduces the computational overhead associated with multi-source feature selection but also significantly enhances the efficiency as the volume and variety of data sources increase.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":"309 ","pages":"Article 112781"},"PeriodicalIF":7.2,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142758825","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Dual-decoding branch contrastive augmentation for image manipulation localization

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-28 DOI: 10.1016/j.knosys.2024.112776

Qiang Zeng, Hongxia Wang, Yang Zhou, Rui Zhang

{"title":"Dual-decoding branch contrastive augmentation for image manipulation localization","authors":"Qiang Zeng, Hongxia Wang, Yang Zhou, Rui Zhang","doi":"10.1016/j.knosys.2024.112776","DOIUrl":"10.1016/j.knosys.2024.112776","url":null,"abstract":"<div><div>With the rapid development of image editing techniques, forensic analysis to detect malicious image manipulation has become an important research topic. The current image manipulation detection and localization methods can accommodate diverse forms of tampering. However, their approach to handling various tampering detection types is limited to a uniform regression pathway. This approach fails to recognize the unique characteristics of copy-move tampering, which is significantly different from other tampering types. Employing a generic detection methodology indiscriminately poses the risk of confusing the training regression trajectory of the deep learning models. To mitigate this challenge, this paper introduces a novel framework featuring a dual-decoding branch structure specifically designed to augment features pertinent to copy-move tampering types. Moreover, it facilitates the detection of tampered regions, irrespective of the tampering type, within the main branch. To achieve this goal, we first introduce a contrastive augmentation module in the encoder, which maximizes the feature space distance between the manipulation regions and pristine regions. Next, we design a parallel attention module to extract more diverse multiscale features. Moreover, we introduce a constrained shifted-window dual attention module to extract tampering noise features. In the decoder, we design a dual-decoding branch to capture both the homologous and tampering features, and we employ contrastive learning to minimize the feature space distance of the homologous regions for copy-move manipulation detection. Finally, we design a category normalization loss function to balance the model’s attention across each category. Extensive experiments demonstrate that the proposed approach achieves state-of-the-art performance on various benchmark datasets.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":"309 ","pages":"Article 112776"},"PeriodicalIF":7.2,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142759454","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

EMGE: Entities and Mentions Gradual Enhancement with semantics and connection modelling for document-level relation extraction

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-28 DOI: 10.1016/j.knosys.2024.112777

Guojun Chen , Panfeng Chen , Qi Wang , Hui Li , Xin Zhou , Xibin Wang , Aihua Yu , Xingzhi Deng

{"title":"EMGE: Entities and Mentions Gradual Enhancement with semantics and connection modelling for document-level relation extraction","authors":"Guojun Chen , Panfeng Chen , Qi Wang , Hui Li , Xin Zhou , Xibin Wang , Aihua Yu , Xingzhi Deng","doi":"10.1016/j.knosys.2024.112777","DOIUrl":"10.1016/j.knosys.2024.112777","url":null,"abstract":"<div><div>Relation extraction is the process of identifying connections between entities in unstructured text and is a critical component of entity-centred information extraction to uncover latent knowledge structures in complex documents. Although graph-based methods have pushed the state-of-the-art forward in relation extraction, current approaches still exhibit limitations. These include incomplete capture of graph structural features, inadequate modelling of long-distance dependencies and imprecise representation of complex entity interactions. A novel <em>E</em>ntities and <em>M</em>entions <em>G</em>radual <em>E</em>nhancement framework called <em>EMGE</em> is proposed. It integrates both contextual and structural information to robustly enhance entity representations for document-level relation extraction. It comprises three primary components: 1) a dynamic relation aware enhancement mechanism to comprehensively encode graph structural features; 2) a multi-scale feature enhancement module to effectively capture long-distance dependencies; and 3) an entity-mention pair enhancement mechanism to yield precise representations of classification targets. Extensive empirical evaluation on five widely-adopted datasets demonstrates that <em>EMGE</em> achieves promising performance. Particularly noteworthy are the substantial gains obtained on the challenging CDR dataset, where <em>EMGE</em> achieved relative improvements of 1.5%, 8.8%, and 3.5% over the strongest baseline in terms of the Intra-F1, Inter-F1 and Overall-F1 metrics, respectively. Further experimental results demonstrate that the proposed model outperforms the popular large language model in relation extraction tasks. Our code is available on github. <span><span><sup>1</sup></span></span></div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":"309 ","pages":"Article 112777"},"PeriodicalIF":7.2,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142759455","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Explainable next POI recommendation based on spatial–temporal disentanglement representation and pseudo profile generation

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-28 DOI: 10.1016/j.knosys.2024.112784

Jun Zeng, Hongjin Tao, Junhao Wen, Min Gao

{"title":"Explainable next POI recommendation based on spatial–temporal disentanglement representation and pseudo profile generation","authors":"Jun Zeng, Hongjin Tao, Junhao Wen, Min Gao","doi":"10.1016/j.knosys.2024.112784","DOIUrl":"10.1016/j.knosys.2024.112784","url":null,"abstract":"<div><div>The current research in Point-of-Interest (POI) recommendation primarily aims to decipher users’ transitional patterns to predict their future location visits. Traditional approaches often intertwine various features to model these check-in transitions, which inadvertently compromises the quality of the resulting representations. This issue is compounded in both industrial and academic settings, where user-generated textual data is frequently inaccessible or restricted due to privacy concerns. Such limitations in user profiles pose significant challenges to the effectiveness of subsequent applications. In response to these challenges, the recent rise of Large Language Models (LLMs) offers a novel perspective. Diverging from the conventional approach of leveraging LLMs for semantic-based next check-in predictions, our research investigates the potential of integrating LLMs with sequential recommendation systems. This integration aims to augment feature dimensions and facilitate the generation of explicit explanations. To this end, we introduce CrossDR-Gen, a Cross-sequence Location Disentanglement Representation methodology. CrossDR-Gen is specifically designed for next POI recommendation and explanation generation. It uniquely considers spatial and temporal factors in shaping check-in behaviors, offering a comprehensive global view of location transitions. Crucially, CrossDR-Gen utilizes LLMs for pseudo profile generation in scenarios with limited semantic context, thereby enriching user features without relying on additional textual profiles or conversational data. Our experiments on real-world datasets demonstrate that CrossDR-Gen not only excels in addressing cold-start scenarios but also showcases robust recommendation capabilities. These findings validate the effectiveness of our proposed cooperative paradigm between LLMs and sequential recommendation models, highlighting a promising avenue for future research in POI recommendation systems.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":"309 ","pages":"Article 112784"},"PeriodicalIF":7.2,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142758848","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0