相关参考文献
注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article
Computer Science, Artificial Intelligence
Xin Liu et al.
Summary: This study proposes a real-time preference mining model (RTPM) based on LSTM for recommending users the next POI with time restrictions. The model mines users' real-time preferences from long-term and short-term preferences, and designs a category filter at the recommendation stage to improve accuracy by filtering out unpopular POIs.
Article
Computer Science, Artificial Intelligence
Yongheng Liu et al.
Summary: POI recommendation is crucial in location-based social networks, utilizing user-generated check-in data to personalize recommendations. Understanding spatial and temporal effects is essential for users' decision-making. The STORE model, incorporating spatiotemporal effects analysis, outperforms existing methods in recommendation performance.
APPLIED INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Yuxia Wu et al.
Summary: Next POI recommendation has been extensively studied, and this article proposes a novel method called Personalized Long- and Short-term Preference Learning (PLSPL) to model users' specific preferences. The method considers both users' general taste and recent sequential behaviors, as well as the different influences of locations and categories of POIs on user preference.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2022)
Article
Computer Science, Artificial Intelligence
Sajal Halder et al.
Summary: Personalized itinerary recommendation is a widely studied area, but existing solutions have some issues. To address these issues, this research proposes an adaptive Monte Carlo tree search (MCTS)-based reinforcement learning algorithm called EffiTourRec. The algorithm considers multiple factors when selecting POIs and reduces non-optimal and duplicated itinerary generation through an efficient search pruning technique.
KNOWLEDGE AND INFORMATION SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Juan Ni et al.
Summary: Recent studies have shown that attention mechanisms are crucial for accurately capturing user interests in recommender systems. The proposed CCDMA model achieves higher accuracy in extracting user and item latent feature vectors, considering both self-attention and cross-attention, and optimizing a comparative learning framework, leading to significant improvements in various evaluation metrics.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Mehri Davtalab et al.
Summary: The study proposed a social spatio-temporal probabilistic matrix factorization (SSTPMF) model that integrates various spaces and utilizes POI similarity and user similarity for point of interest (POI) recommendation. The model showed improved recommendation performance by considering POI correlation and user similarity, and effectively alleviated the cold start problem compared to state-of-the-art methods on two real data sets, Foursquare and Gowalla.
KNOWLEDGE AND INFORMATION SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Guixiang Zhu et al.
Summary: This study introduces a novel model named Neural Attentive Travel package Recommendation (NATR) for tourism e-commerce, combining users' long-term and short-term preferences through travel package encoder and user encoder. The model demonstrates significant performance advantages over competitive methods in experiments.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Giannis Christoforidis et al.
Summary: The rapid growth of users' involvement in Location-Based Social Networks has led to the challenge of accessing and retrieving relevant information close to users' preferences. The proposed unified model, RELINE, addresses this challenge by jointly learning user and point-of-interest dynamics. By embedding multiple relational graphs into a shared latent space, RELINE captures social, geographical, temporal, and preference dynamics to provide accurate and personalized recommendations. Performance evaluations demonstrate significant improvements over existing state-of-the-art methods.
KNOWLEDGE AND INFORMATION SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Lin Liu et al.
Summary: Session-based recommendation aims to predict next items based on users' anonymous behavior sequence within a short time. This study proposes a novel method CaSe4SR that utilizes category sequence graph to augment session-based recommendation, which outperforms other state-of-the-art methods consistently. The category sequence graph is beneficial for next-item recommendation in sessions with different lengths.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Tipajin Thaipisutikul et al.
Summary: This paper proposes a novel Context-aware Recommender System based on a Deep Sequential Learning approach to capture users' dynamic preferences by modeling hierarchical relationships between contexts and items, achieving significant improvement over state-of-the-art methods in experiments.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Hang Zhang et al.
Summary: This paper introduces a novel POI recommendation framework named U-CF-Memory-Stickiness, which emphasizes the importance of personal psychological effects and memory-based preference evolution in human mobility patterns, as well as the consideration of POI stickiness based on visiting frequency to identify important POIs reflecting stable interests. By incorporating memory-based preferences and POI stickiness into a user-based collaborative filtering framework, the proposed method outperforms existing methods in POI recommendations, as demonstrated in experiments on a real LBSN dataset.
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
(2021)
Article
Computer Science, Information Systems
Honglian Wang et al.
Summary: This paper proposes a new method for next point-of-interest (POI) recommendation, called DSPR, by exploring user preferences and real-time demand simultaneously to support the final POI recommendation. Experimental results show that DSPR outperforms many state-of-the-art methods in recommendation performance.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Jiyong Zhang et al.
Summary: This work proposes a generic POI recommendation framework GNN-POI, leveraging Graph Neural Networks (GNNs) to improve POI recommendation by learning node representations, which significantly outperforms existing models in terms of precision, recall, and Normalized Discounted Cumulative Gain (NDCG) based on extensive experiments over three real LBSN datasets.
Article
Computer Science, Artificial Intelligence
Meihui Shi et al.
Summary: This study proposes a novel time-aware POI recommendation method based on an attentional memory network with correlation-based embedding. By capturing geographical influence and micro-level relationships in time slots, the method has shown significant improvements in recommendation accuracy.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Information Systems
Liwei Huang et al.
Summary: The concept of next point-of-interest (POI) recommendation, focusing on predicting user's next destination by learning sequential patterns of check-in behavior, has been proposed. Researchers introduced two new networks, ST-LSTM and ATST-LSTM, to improve the performance of next POI recommendation. Experimental results showed that these new approaches outperformed existing methods in terms of performance evaluation metrics.
IEEE TRANSACTIONS ON SERVICES COMPUTING
(2021)
Article
Computer Science, Artificial Intelligence
Lei Chen et al.
Summary: This paper proposes an AMT-IRE framework, which dynamically learns the inner relations between group members and obtains consensus group preferences via the attention mechanism. By integrating POI categories and textual information, combined with attention networks, group itineraries are recommended effectively.
KNOWLEDGE AND INFORMATION SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Zahra Bahari Sojahrood et al.
Summary: Group recommendation using location-based social networks has attracted research attention. This paper proposes a method for point of interest (POI) group recommendation that considers user influence modeled fuzzily, historical check-in data, and factors like category, distance, and time. Experimental results show improved accuracy in POI group recommendations, especially when user influence is calculated using the fuzzy approach. The study also reveals differences in user behavior when choosing places to visit alone or in a group.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Jianxing Zheng et al.
Summary: This paper proposes an attentional attribute and interaction method and constructs a type-specific matrix to learn user and item representations using heterogeneous type-specific information. The approach not only addresses the sparsity problem of user-item interactions in recommender systems but also predicts the rating relationship of the nodes through a translation mechanism.
INFORMATION PROCESSING & MANAGEMENT
(2021)
Article
Computer Science, Artificial Intelligence
Qigang Liu et al.
Summary: This study focuses on studying the representation and mining of user preference from check-in data for POI recommendation. A new negative sampling method is proposed for enhancing the quality of training data. Experimental results show that the proposed approach outperforms state-of-the-art models.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Fan Zhou et al.
Summary: This study introduces a Self-supervised Mobility Learning framework for sparse and noisy human mobility data, focusing on enhancing trajectory representations with rich spatio-temporal contexts and augmented traces. Contrastive instance discrimination is first introduced to improve model training accuracy by distinguishing real user check-ins from negative samples.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Cybernetics
Chenwang Zheng et al.
Summary: The paper proposes a memory augmented hierarchical attention network (MAHAN), which addresses the challenges in next point-of-interest recommendation by considering both short-term check-in sequences and long-term memories. They design a spatiotemporal self-attention network (ST-SAN) to capture users' complex interest tendencies in the short-term, and use a memory network for long-term preferences modeling. Additionally, they integrate the ST-SAN and memory network using a coattention network/mechanism to fully learn the dynamic interaction between long- and short-term preferences.
IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Tongcun Liu et al.
Article
Computer Science, Information Systems
Yaxue Ma et al.
INFORMATION PROCESSING & MANAGEMENT
(2020)
Article
Computer Science, Artificial Intelligence
Yingxue Ma et al.
Article
Computer Science, Information Systems
Ting Zhong et al.
WORLD WIDE WEB-INTERNET AND WEB INFORMATION SYSTEMS
(2020)
Article
Computer Science, Artificial Intelligence
Guangyao Pang et al.
APPLIED SOFT COMPUTING
(2020)
Article
Computer Science, Information Systems
Mingxin Gan et al.
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
(2019)
Proceedings Paper
Computer Science, Information Systems
Xiang Wang et al.
PROCEEDINGS OF THE 42ND INTERNATIONAL ACM SIGIR CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION RETRIEVAL (SIGIR '19)
(2019)
Proceedings Paper
Computer Science, Artificial Intelligence
Ranzhen Li et al.
2018 IEEE INTERNATIONAL CONFERENCE ON DATA MINING (ICDM)
(2018)
Article
Computer Science, Artificial Intelligence
Hongzhi Yin et al.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2017)
Article
Computer Science, Artificial Intelligence
Xingyi Ren et al.