Successful Recommender System Research Articles

A Collaborative Filtering Recommendation Algorithm Based on Product Clustering

E-commerce recommendation system is one of the most important and the most successful application field of information intelligent technology. Recommender systems help to overcome the problem of information overload on the Internet by providing personalized recommendations to the customers. Recommendation algorithm is the core of the recommendation system. Collaborative filtering recommendation algorithm is the personalized recommendation algorithm that is used widely in e-commerce recommendation system. Collaborative filtering has been a comprehensive approach in recommendation system. But data are always sparse. This becomes the bottleneck of collaborative filtering. Collaborative filtering is regarded as one of the most successful recommender systems within the last decade, which predicts unknown ratings by analyzing the known ratings. In this paper, an electronic commerce collaborative filtering recommendation algorithm based on product clustering is given. In this approach, the clustering of product is used to search the recommendation neighbors in the clustering centers.

Maximizing customer satisfaction through an online recommendation system: A novel associative classification model

Offering online personalized recommendation services helps improve customer satisfaction. Conventionally, a recommendation system is considered as a success if clients purchase the recommended products. However, the act of purchasing itself does not guarantee satisfaction and a truly successful recommendation system should be one that maximizes the customer's after-use gratification. By employing an innovative associative classification method, we are able to predict a customer's ultimate pleasure. Based on customer's characteristics, a product will be recommended to the potential buyer if our model predicts his/her satisfaction level will be high. The feasibility of the proposed recommendation system is validated through laptop Inspiron 1525.

Users’ needs, desires, and design preferences for recommendations in the living room

By using a combination of playful probing, creative cultural probing and technology probing, user needs and desires for recommendations in the living room were investigated. 40 households with a total of 126 participants took part in the ethnographic study. The participants used the probing material, received recommendations from an interactive TV (iTV) recommender system, and logged their TV watching behaviour. The findings show that recommendations should preferably be given within the same media (the iTV system), that users prefer to influence the proposal that they receive rather than getting automated recommendations and that the design should ideally support recommendations for the whole household and additionally for individuals. The study revealed that in order to understand which factors account for a successful recommendation system from the user’s perspective, especially trust and security have to be further investigated.

COLLABORATIVE FILTERING FOR MULTI-CLASS DATA USING BAYESIAN NETWORKS

As one of the most successful recommender systems, collaborative filtering (CF) algorithms are required to deal with high sparsity and high requirement of scalability amongst other challenges. Bayesian networks (BNs), one of the most frequently used classifiers, can be used for CF tasks. Previous works on applying BNs to CF tasks were mainly focused on binary-class data, and used simple or basic Bayesian classifiers.1,2 In this work, we apply advanced BNs models to CF tasks instead of simple ones, and work on real-world multi-class CF data instead of synthetic binary-class data. Empirical results show that with their ability to deal with incomplete data, the extended logistic regression on tree augmented naïve Bayes (TAN-ELR)3 CF model consistently performs better than the traditional Pearson correlation-based CF algorithm for the rating data that have few items or high missing rates. In addition, the ELR-optimized BNs CF models are robust in terms of the ability to make predictions, while the robustness of the Pearson correlation-based CF algorithm degrades as the sparseness of the data increases.

A Movie E-shop Recommendation Model Based on Web Usage and Ontological Data

Recommendation systems are special personalization tools that help users to find interesting information and services in complex online shops. Even though today’s e-commerce environments have drastically evolved and now incorporate techniques from other domains and application areas such as Web mining, semantics, arti- ficial intelligence, user modeling, and profiling setting up a successful recommendation system is not a trivial or straightforward task. This paper argues that by monitoring, analyzing, and understanding the behavior of customers, their demographics, opinions, preferences, and history, as well as taking into consideration the specific e-shop ontology and by applying Web mining techniques, the effectiveness of produced recommendations can be significantly improved. In this way, the e-shop may upgrade users’ interaction, increase its usability, convert users to buyers, retain current customers, and establish long-term and loyal one-to-one relationships.