Mode of access: World Wide Web.

System requirements: Adobe Acrobat Reader.

Part of: Synthesis digital library of engineering and computer science.

Series from website.

Includes bibliographical references (pages 161-179).

1. The recommendation process -- 1.1 Introduction -- 1.2 Formal framework -- 1.2.1 Evaluation -- 1.2.2 Main challenges -- 1.3 Recommendation as information filtering -- 1.3.1 Demographic filtering -- 1.3.2 Content-based filtering -- 1.4 Collaborative filtering -- 1.4.1 Neighborhood-based approaches -- 1.4.2 Latent factor models -- 1.4.3 Baseline models and collaborative filtering --

2. Probabilistic models for collaborative filtering -- 2.1 Predictive modeling -- 2.2 Mixture membership models -- 2.2.1 Mixtures and predictive modeling -- 2.2.2 Model-based co-clustering -- 2.3 Probabilistic latent semantic models -- 2.3.1 Probabilistic latent semantic analysis -- 2.3.2 Probabilistic matrix factorization -- 2.4 Summary --

3. Bayesian modeling -- 3.1 Bayesian regularization and model selection -- 3.2 Latent Dirichlet allocation -- 3.2.1 Inference and parameter estimation -- 3.2.2 Bayesian topic models for recommendation -- 3.3 Bayesian co-clustering -- 3.3.1 Hierarchical models -- 3.4 Bayesian matrix factorization -- 3.5 Summary --

4. Exploiting probabilistic models -- 4.1 Probabilistic modeling and decision theory -- 4.1.1 Minimizing the prediction error -- 4.1.2 Recommendation accuracy -- 4.2 Beyond prediction accuracy -- 4.2.1 Data analysis with topic models -- 4.2.2 Pattern discovery using co-clusters -- 4.2.3 Diversification with topic models --

5. Contextual information -- 5.1 Integrating content features -- 5.1.1 The cold-start problem -- 5.1.2 Modeling text and preferences -- 5.2 Sequential modeling -- 5.2.1 Markov models -- 5.2.2 Probabilistic tensor factorization --

6. Social recommender systems -- 6.1 Modeling social rating networks -- 6.2 Probabilistic approaches for social rating networks -- 6.2.1 Network-aware topic models -- 6.2.2 Social probabilistic matrix factorization -- 6.2.3 Stochastic block models for social rating networks -- 6.3 Influence in social networks -- 6.3.1 Identifying social influence -- 6.3.2 Influence maximization and viral marketing -- 6.3.3 Exploiting influence in recommender systems --

7. Conclusions -- 7.1 Application-specific challenges -- 7.2 Technological challenges --

A. Parameter estimation and inference -- A1. The expectation maximization algorithm -- A2. Variational inference -- A3. Gibbs sampling -- Bibliography -- Authors' biographies.

Abstract freely available; full-text restricted to subscribers or individual document purchasers.

Compendex

INSPEC

Google scholar

Google book search

The importance of accurate recommender systems has been widely recognized by academia and industry, and recommendation is rapidly becoming one of the most successful applications of data mining and machine learning. Understanding and predicting the choices and preferences of users is a challenging task: real-world scenarios involve users behaving in complex situations, where prior beliefs, specific tendencies, and reciprocal influences jointly contribute to determining the preferences of users toward huge amounts of information, services, and products. Probabilistic modeling represents a robust formal mathematical framework to model these assumptions and study their effects in the recommendation process.

Also available in print.

Title from PDF title page (viewed on June 20, 2014).