Abstract:Web portal services have become an important medium to deliver digital content (e.g. news, advertisements, etc.) to Web users in a timely fashion. To attract more users to various content modules on the Web portal, it is necessary to design a recommender system that can effectively achieve Web portal content optimization by automatically estimating content item attractiveness and relevance to user interests. The state-of-the-art online learning methodology adapts dedicated pointwise models to independently estimate the attractiveness score for each candidate content item. Although such pointwise models can be easily adapted for online recommendation, there still remain a few critical problems. First, this pointwise methodology fails to use invaluable user preferences between content items. Moreover, the performance of pointwise models decreases drastically when facing the problem of sparse learning samples. To address these problems, we propose exploring a new dynamic pairwise learning methodology for Web portal content optimization in which we exploit dynamic user preferences extracted based on users' actions on portal services to compute the attractiveness scores of content items. In this article, we introduce two specific pairwise learning algorithms, a straightforward graph-based algorithm and a formalized Bayesian modeling one. Experiments on large-scale data from a commercial Web portal demonstrate the significant improvement of pairwise methodologies over the baseline pointwise models. Further analysis illustrates that our new pairwise learning approaches can benefit personalized recommendation more than pointwise models, since the data sparsity is more critical for personalized content optimization.

Mining User Preferences for Recommendation: A Competition Perspective

Real-time Recommendation Method Based on Browsing Preferences Mining

Improved Metric-Based Recommender by Historical Interactions.

User Preference Quantity Versus Recommendation Performance: A Preliminary Study

Using Quantitative Association Rules in Collaborative Filtering

Mining User/Movie Preferred Features Based on Reviews for Video Recommendation System

Collaborative Filtering with Short Term Preferences Mining.

SOCIAL INTEREST FOR USER SELECTING ITEMS IN RECOMMENDER SYSTEMS

Mining Preferences from Superior and Inferior Examples

Employing F-MADM to Derive User Preference Model from Item Features and Rating Information for Personalized Recommendation

Personalized Recommendation Combining User Interest and Social Circle

Finding What Users Look for by Attribute-Aware Personalized Item Comparison in Relevant Recommendation.

Exploiting User Preference for Online Learning in Web Content Optimization Systems

Bridging User Interest to Item Content for Recommender Systems: An Optimization Model

Mining Interest Trends and Adaptively Assigning SampleWeight for Session-based Recommendation

Combining User Preferences and User Opinions for Accurate Recommendation

What You Like, What I Am: Online Dating Recommendation Via Matching Individual Preferences with Features

Personalized Recommendation Algorithm Based on Preference Features

Incorporating User Preferences Across Multiple Topics Into Collaborative Filtering For Personalized Merchant Recommendation

Mining context-related sequential patterns for recommendation systems

Mining Mobile User Preferences for Personalized Context-Aware Recommendation.