Lu Yu

Jifei Luo, Wenzheng Wu, Hantao Yao et al.

Diffusion-based re-ranking methods are effective in modeling the data manifolds through similarity propagation in affinity graphs. However, positive signals tend to diminish over several steps away from the source, reducing discriminative power beyond local regions. To address this issue, we introduce the Locality Preserving Markovian Transition (LPMT) framework, which employs a long-term thermodynamic transition process with multiple states for accurate manifold distance measurement. The proposed LPMT first integrates diffusion processes across separate graphs using Bidirectional Collaborative Diffusion (BCD) to establish strong similarity relationships. Afterwards, Locality State Embedding (LSE) encodes each instance into a distribution for enhanced local consistency. These distributions are interconnected via the Thermodynamic Markovian Transition (TMT) process, enabling efficient global retrieval while maintaining local effectiveness. Experimental results across diverse tasks confirm the effectiveness of LPMT for instance retrieval.

Lu Yu, Junming Huang, Chuang Liu et al.

Interactions between search and recommendation have recently attracted significant attention, and several studies have shown that many potential applications involve with a joint problem of producing recommendations to users with respect to a given query, termed $Collaborative$ $Retrieval$ (CR). Successful algorithms designed for CR should be potentially flexible at dealing with the sparsity challenges since the setup of collaborative retrieval associates with a given $query$ $\times$ $user$ $\times$ $item$ tensor instead of traditional $user$ $\times$ $item$ matrix. Recently, several works are proposed to study CR task from users' perspective. In this paper, we aim to sufficiently explore the sophisticated relationship of each $query$ $\times$ $user$ $\times$ $item$ triple from items' perspective. By integrating item-based collaborative information for this joint task, we present an alternative factorized model that could better evaluate the ranks of those items with sparse information for the given query-user pair. In addition, we suggest to employ a recently proposed scalable ranking learning algorithm, namely BPR, to optimize the state-of-the-art approach, $Latent$ $Collaborative$ $Retrieval$ model, instead of the original learning algorithm. The experimental results on two real-world datasets, (i.e. \emph{Last.fm}, \emph{Yelp}), demonstrate the efficiency and effectiveness of our proposed approach.

Lu Yu, Chuang Liu, Zi-Ke Zhang

Recommender systems, which can significantly help users find their interested items from the information era, has attracted an increasing attention from both the scientific and application society. One of the widest applied recommendation methods is the Matrix Factorization (MF). However, most of MF based approaches focus on the user-item rating matrix, but ignoring the ingredients which may have significant influence on users' preferences on items. In this paper, we propose a multi-linear interactive MF algorithm (MLIMF) to model the interactions between the users and each event associated with their final decisions. Our model considers not only the user-item rating information but also the pairwise interactions based on some empirically supported factors. In addition, we compared the proposed model with three typical other methods: user-based collaborative filtering (UCF), item-based collaborative filtering (ICF) and regularized MF (RMF). Experimental results on two real-world datasets, \emph{MovieLens} 1M and \emph{MovieLens} 100k, show that our method performs much better than other three methods in the accuracy of recommendation. This work may shed some light on the in-depth understanding of modeling user online behaviors and the consequent decisions.

Chu-Xu Zhang, Zi-Ke Zhang, Lu Yu et al.

The past few years have witnessed the great success of recommender systems, which can significantly help users find out personalized items for them from the information era. One of the most widely applied recommendation methods is the Matrix Factorization (MF). However, most of researches on this topic have focused on mining the direct relationships between users and items. In this paper, we optimize the standard MF by integrating the user clustering regularization term. Our model considers not only the user-item rating information, but also takes into account the user interest. We compared the proposed model with three typical other methods: User-Mean (UM), Item-Mean (IM) and standard MF. Experimental results on a real-world dataset, \emph{MovieLens}, show that our method performs much better than other three methods in the accuracy of recommendation.