Two-Way Latent Grouping Model for User Preference Prediction
نویسندگان
چکیده
We introduce a novel latent grouping model for predicting the relevance of a new document to a user. The model assumes a latent group structure for both users and documents. We compared the model against a state-of-the-art method, the User Rating Profile model, where only users have a latent group structure. We estimate both models by Gibbs sampling. The new method predicts relevance more accurately for new documents that have few known ratings. The reason is that generalization over documents then becomes necessary and hence the twoway grouping is profitable.
منابع مشابه
Probabilistic Canonical Tensor Decomposition for Predicting User Preference
We propose a model to infer a user’s preference ranking over musicians from a sparse set of pairwise preferences of the form ”user k prefers artist i over artist j”. The goal is to approximate the data with a low-rank factor model using canonical tensor decomposition. A user-specific pairwise preference is modeled as the sign of a 3-way tensor inner product of latent factor vectors associated w...
متن کاملA Context-Aware User-Item Representation Learning for Item Recommendation
Both reviews and user-item interactions (i.e., rating scores) have been widely adopted for user rating prediction. However, these existing techniques mainly extract the latent representations for users and items in an independent and static manner. That is, a single static feature vector is derived to encode her preference without considering the particular characteristics of each candidate ite...
متن کاملTwo-Way Grouping by One-Way Topic Models
We tackle the problem of new users or documents in collaborative filtering. Generalization over users by grouping them into user groups is beneficial when a rating is to be predicted for a relatively new document having only few observed ratings. The same applies for documents in the case of new users. We have shown earlier that if there are both new users and new documents, two-way generalizat...
متن کاملHierarchical Latent Factors for Preference Data
In this work we propose a probabilistic hierarchical generative approach for users’ preference data, which is designed to overcome the limitation of current methodologies in Recommender Systems and thus to meet both prediction and recommendation accuracy. The Bayesian Hierarchical User Community Model (BH-UCM) focuses both on modeling the popularity of items and the distribution over item ratin...
متن کاملBalancing Prediction and Recommendation Accuracy: Hierarchical Latent Factors for Preference Data
Recent works in Recommender Systems (RS) have investigated the relationships between the prediction accuracy, i.e. the ability of a RS to minimize a cost function (for instance the RMSE measure) in estimating users’ preferences, and the accuracy of the recommendation list provided to users. State-of-the-art recommendation algorithms, which focus on the minimization of RMSE, have shown to achiev...
متن کامل