Temporal recommendation on graphs via long- and short-term preference fusion

By Liang Xiang, Quan Yuan, Shiwan Zhao, Li Chen, Xiatian Zhang, Qing Yang, Jimeng Sun. 2010

410 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems, Management Science and Operations Research

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

In Management Science and Operations Research

Citations in the corpus

From amateurs to connoisseurs: modeling the evolution of user expertise through online reviews (2013, background): “due to external events), and long-term trends [40].”
Daily-Aware Personalized Recommendation based on Feature-Level Time Series Analysis (2015, background): “[34] further extended the dynamic modeling to implicit feedback through random walk on sessions-based temporal graphs; Lu [19] and Shi [25] adapted the famous Matrix Factorization (MF) [28] approaches for time-aware CF, while Karatzoglou [15] and Gantner [13] leveraged tensor factorization to integr”
TriRank: Review-aware Explainable Recommendation by Modeling Aspects (2015, methodology): “A state-of-the-art for top-N recommendation, which performs Singular Value Decomposition on the whole matrix, thus directly considering all instances. unlike other latent factor methods that optimize against error only on rated instances.”
Recommendations : Modeling Music Ratings with Temporal Dynamics and Item Taxonomy (2011, background): “Thus, a few recent works [13, 26] highlighted the significance of”
NAIS: Neural Attentive Item Similarity Model for Recommendation (2018, methodology): “Both metrics have been widely used to evaluate top-K recommendation [1], [32] and ranking systems [33] in information retrieval literatures.”
CoNet: Collaborative Cross Networks for Cross-Domain Recommendation (2018, background): “Random walk and heterogeneous networks are adapted for recommendation [37, 42].”
Visually Explainable Recommendation (2018, background): “Top-N • Hit-Ratio (HR): Hit-ratio gives the percentage of users that can receive at least one correct recommendation, which has been widely used in previous work [15, 33].”
Sequence-Aware Recommender Systems (2018, background): “[112] focus on context adaption to detect short-term user intents and interests.”
Recommending Groups to Users Using User-Group Engagement and Time-Dependent Matrix Factorization (2016, methodology): “Other authors have examined time-dependent methods such as tensor factorization, session-based temporal graph model and dynamic matrix factorization etc. (Xiong et al. 2010; Xiang et al. 2010; Koenigstein, Dror, and Koren 2011; Lathia, Hailes, and Capra 2009; Chua, Oentaryo, and Lim 2013).”
Recommender Systems (2016, positive): “Temporal recommendations with the use of random walks on graphs are discussed in [639].”
Learning Hierarchical Representation Model for NextBasket Recommendation (2015, methodology): “• Hit-Ratio: Hit-Ratio is a All-but-One measure used in recommendation [13, 28].”
Recommender Systems (2012, methodology): “It is an ongoing line of research whether and how value of old opinions should decay with time and what are the typical temporary patterns in user evaluations and item relevance [41, 42].”
Next-item Recommendation with Sequential Hypergraphs (2020, positive): “To model the temporally dynamic patterns and predict for future behaviors, Session-based Temporal Graph (STG) [42] is proposed to connect users, items and sessions in a graph.”
Sequential Recommendation with User Memory Networks (2018, methodology): “…global average for better interpretability [16]: • Hit-ratio (HR): Hit-ratio gives the percentage of users that can receive at least one correct recommendation, which has been widely used in previous work [16, 30]: where I ( x ) is an indicator function whose value is 1 when x > 0, and 0 otherwise.”
Path Problems in Temporal Graphs (2014, methodology): “In [20], a temporal graph is used to model users’ long-term and short-term preferences over time and the temporal information is used for recommendation.”
Time-aware point-of-interest recommendation (2013, methodology): “In other domains time factor is considered for recommendation in several studies, including using item-based CF method [7], matrix factorization method [11] and random walk based method [19].”
Collaborative Filtering beyond the User-Item Matrix (2014, methodology): “Temporal recommendations are mined from the graph using random walk with restarts [Xiang et al. 2010].”
Time-aware recommender systems: a comprehensive survey and analysis of existing evaluation protocols (2013, background): “It facilitates tracking the evolution of user preferences (Xiang et al. 2010), enabling e.g. to identify periodicity in user habits and interests (Campos et al. 2011a), which is a key input for many context-aware RS (CARS) (Baltrunas and Amatriain 2009; Panniello et al. 2009b).”