Article
Computer Science, Information Systems
Iwao Tanuma, Tomoko Matsui
Summary: This study proposes a hybrid recommendation method based on collaborative filtering, which models the number of interactions as a Poisson-distributed content information generation process using a variational autoencoder. The method effectively utilizes content information for predicting interactions and helps to overcome the cold-start problem.
Article
Computer Science, Artificial Intelligence
Yi-Hung Liu, Yen-Liang Chen, Po-Ya Chang
Summary: With the increasing number of mobile applications, it has become difficult for users to find the most suitable and interesting ones. This study proposes a better model for mobile app recommendation by combining matrix factorization, user reviews, and deep learning methods. Experimental results show that this model outperforms existing methods.
DECISION SUPPORT SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Weina Zhang, Xingming Zhang, Dongpei Chen
Summary: Implicit feedback data has various forms of interaction, such as clicking, collection, and play count, posing a challenge to recommendation systems. This paper introduces a Causal Neural Fuzzy Inference algorithm to address missing data in implicit recommendations through joint learning, demonstrating effectiveness and advancement in experiments on realistic datasets.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Xiaodong Feng, Zhen Liu, Wenbing Wu, Wenbo Zuo
Summary: The rapid development of social recommendation in recent years has greatly improved the performance of recommender systems, especially for the cold start problem. However, existing techniques based on matrix factorization do not effectively capture the complex nonlinear relationships between users and items, as well as between users themselves. To address this, deep learning is employed to model the social network-enhanced collaborative filtering problem. By simultaneously modeling the social and item domain interactions, the proposed SoNeuMF framework shows significant improvements in recommendation accuracy compared to state-of-the-art methods, as demonstrated by comprehensive experiments on real-world datasets.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Mathematics
Yongheng Mu, Yun Wu
Summary: Recommendation systems are widely used to provide personalized content and services to users efficiently. In this paper, a personalized multimodal movie recommendation system based on deep learning and multimodal data analysis was proposed. Real-world MovieLens datasets were used to test the effectiveness of the algorithm, which achieved improved accuracy in predicting movie scores compared to traditional collaborative filtering approaches. The combination of deep learning and multimodal data analysis can help alleviate the sparse data problem and enhance the performance of recommendation systems.
Article
Chemistry, Multidisciplinary
Jibing Gong, Xinghao Zhang, Qing Li, Cheng Wang, Yaxi Song, Zhiyong Zhao, Shuli Wang
Summary: By combining display information with implicit information, a new top-N recommendation method called MFDNN was proposed for Heterogeneous Information Networks (HINs), significantly improving hit ratio and normalized discounted cumulative gain in experiments.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Artificial Intelligence
Xiaoxuan Shen, Baolin Yi, Hai Liu, Wei Zhang, Zhaoli Zhang, Sannyuya Liu, Naixue Xiong
Summary: In this article, we proposed the DVMF framework which combines deep learning and fully Bayesian treatment recommendation techniques to efficiently integrate various auxiliary information and outperforms existing recommendation algorithms in terms of prediction accuracy in experiments.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2021)
Article
Computer Science, Hardware & Architecture
Shangshang Xu, Haiyan Zhuang, Fuzhen Sun, Shaoqing Wang, Tianhui Wu, Jiawei Dong
Summary: This paper proposes a hybrid method based on probabilistic matrix factorization and directed trust to improve the performance of recommender systems, addressing the sparsity of trust matrix and capturing trust relations among users. Experimental results demonstrate that the proposed algorithm outperforms existing benchmark algorithms.
COMPUTERS & ELECTRICAL ENGINEERING
(2021)
Article
Computer Science, Information Systems
Chao Duan, Jianwen Sun, Kaiqi Li, Qing Li
Summary: The accelerated development of mobile networks and applications has led to exponential expansion of resources, potentially resulting in information overload. One approach to address this issue is recommendation systems, which can provide personalized services. This paper proposes an autoencoder model that utilizes attribute information to improve recommendation efficiency, by extracting user and video features simultaneously while integrating an attention mechanism to generate crucial features.
Article
Computer Science, Artificial Intelligence
Xiaoyao Zheng, Zhen Ni, Xiangnan Zhong, Yonglong Luo
Summary: This article proposes a recommendation model based on kernelized deep neural networks, which encodes the user-item rating matrix and mines implicit relations between users and items to simulate nonlinear user-item interactions, thus improving recommendation performance.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Multidisciplinary Sciences
Connor Bybee, Denis Kleyko, Dmitri E. Nikonov, Amir Khosrowshahi, Bruno A. Olshausen, Friedrich T. Sommer
Summary: This article introduces Ising machines as a prominent approach to solving combinatorial optimization problems on parallel hardware. The authors demonstrate that higher-order Ising machines can be more resource-efficient and provide better solutions compared to traditional second-order Ising machines.
NATURE COMMUNICATIONS
(2023)
Article
Computer Science, Information Systems
Chenyan Zhang, Jing Li, Jia Wu, Donghua Liu, Jun Chang, Rong Gao
Summary: Recommender systems provide effective solutions for handling information overload and have become a popular research topic in both industry and academia. However, existing implicit feedback methods still have some shortcomings, such as biased solutions due to uneven sampling of negative samples and lack of interpretability in recommendation results. Therefore, we propose a new deep recommendation model (DRAT) that utilizes an encoder-decoder structure and adversarial training to provide personalized recommendations. Experimental results show that our proposed model significantly outperforms baseline methods.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTING
(2022)
Article
Computer Science, Artificial Intelligence
Wei Zeng, Ge Fan, Shan Sun, Biao Geng, Weiyi Wang, Jiacheng Li, Weibo Liu
Summary: The deep neural network has been successfully applied to the collaborative filtering problem, capturing side information of users and items and modeling interactions between them. Research trends towards utilizing neural networks with mixed structures to learn better representations, achieving high accuracy with minimal additional computation.
APPLIED SOFT COMPUTING
(2021)
Article
Computer Science, Information Systems
Qianqiao Liang, Xiaolin Zheng, Yan Wang, Mengying Zhu
Summary: This study discusses explainable recommendation systems (ERSs) and the E-3 model in an online learning setting, overcoming the limitations of existing ERSs in offline settings and proposing an online setting to adapt to real-time feedback for improved recommendations and explanations. By immediately integrating online feedback into the model, the E-3 model remarkably enhances performance in online scenarios.
INFORMATION SCIENCES
(2021)
Article
Mathematics, Applied
Hung-Hsu Chou, Carsten Gieshoff, Johannes Maly, Holger Rauhut
Summary: In deep learning, over-parameterization is commonly used and leads to implicit bias. This paper analyzes the dynamics of gradient descent and provides insights into implicit bias. The study also explores time intervals for early stopping and presents empirical evidence for implicit bias in various scenarios.
APPLIED AND COMPUTATIONAL HARMONIC ANALYSIS
(2024)