Presentation | 2010/2/22 Adaptive Design Method for Similarity of Collaborative Filtering based on Optimization Akihiro YAMASHITA, Hidenori KAWAMURA, Keiji SUZUKI, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Collaborative filtering is one of the most popular and effective recommendation algorithms based on user-user or item-item similarity. Generally, distance metrics such as Pearson's correlation coefficient are used as the similarity. Although, effectiveness of similarity computation method was widely discussed from various perspectives, there are few considerations of computation methods for optimal similarity in collaborative filtering. In this research, similarity optimization problem were formulated by defining similarities between a active user and the other users as a vector variable. Then, a quasi-optimal solution was obtained and it was compared with Pearson's correlation coefficient. Additionally, we propose and evaluate similarity computation method based on optimization. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | recommender systems / collaborative filtering / similarity / optimization |
Paper # | AI2009-47 |
Date of Issue |
Conference Information | |
Committee | AI |
---|---|
Conference Date | 2010/2/22(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | |
Vice Chair | |
Secretary | |
Assistant |
Paper Information | |
Registration To | Artificial Intelligence and Knowledge-Based Processing (AI) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Adaptive Design Method for Similarity of Collaborative Filtering based on Optimization |
Sub Title (in English) | |
Keyword(1) | recommender systems |
Keyword(2) | collaborative filtering |
Keyword(3) | similarity |
Keyword(4) | optimization |
1st Author's Name | Akihiro YAMASHITA |
1st Author's Affiliation | Graduate School of Information Science and Technology, Hokkaido University:Japan Society for the Promotion of Science() |
2nd Author's Name | Hidenori KAWAMURA |
2nd Author's Affiliation | Graduate School of Information Science and Technology, Hokkaido University |
3rd Author's Name | Keiji SUZUKI |
3rd Author's Affiliation | Graduate School of Information Science and Technology, Hokkaido University |
Date | 2010/2/22 |
Paper # | AI2009-47 |
Volume (vol) | vol.109 |
Number (no) | 439 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |