Improving Group Recommendations by Identifying Homogenous Subgroups
- Publication Type:
- Conference Proceeding
- Citation:
- Advances in Intelligent Systems and Computing, 2014, 214 pp. 453 - 462
- Issue Date:
- 2014-01-01
Closed Access
Filename | Description | Size | |||
---|---|---|---|---|---|
IGRS_280912.docx | Accepted Manuscript version | 148.89 kB |
Copyright Clearance Process
- Recently Added
- In Progress
- Closed Access
This item is closed access and not available.
Recommender systems have proven their effectiveness in supporting personalised purchasing decisions and e-service intelligence. In order to support members in user groups of recommender systems, recently designed group recommender systems search for data relevant to all group members and discover the agreements between members of online communities. This paper focuses on achieving common satisfaction for groups or communities by, e.g. finding a restaurant for a family or shoes for a group of cheerleaders. It establishes an algorithm, called I-GRS, to devise group recommender systems based on incremental model-based collaborative filtering and applying the Mahalanobis distance and fuzzy membership to create groups of users with similar interests. Finally, an algorithm and related design strategy to build group recommender systems is proposed. A set of experiments is set up to evaluate the performance of the I-GRS algorithm in group recommendations. The results show its effectiveness vis-à-vis the recommendations made by classical recommender systems to single or groups of individuals. © Springer-Verlag Berlin Heidelberg 2014.
Please use this identifier to cite or link to this item: