Presentation | 2010-01-22 Reduction technique of user rating histories for information filtering systems Takayuki UDA, Tetsuo KINOSHITA, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | The information filtering system refers to action histories and evaluation histories of the user to perform filtering. If an operation period becomes long, the value of old evaluation histories deteriorates for the preference changes of the user. In this paper, we propose the reduction technique of user rating histories for information filtering systems. By an evaluation experiment, we confirmed that prediction processing time was shortened without the deterioration of a prediction precision and a coverage. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Collaborative Filtering / Prediction processing time / Recommender System / Prediction Accuracy / Coverage |
Paper # | AI2009-21 |
Date of Issue |
Conference Information | |
Committee | AI |
---|---|
Conference Date | 2010/1/15(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) | Reduction technique of user rating histories for information filtering systems |
Sub Title (in English) | |
Keyword(1) | Collaborative Filtering |
Keyword(2) | Prediction processing time |
Keyword(3) | Recommender System |
Keyword(4) | Prediction Accuracy |
Keyword(5) | Coverage |
1st Author's Name | Takayuki UDA |
1st Author's Affiliation | Secretariat, INSTITUTE of INFORMATION SECURITY() |
2nd Author's Name | Tetsuo KINOSHITA |
2nd Author's Affiliation | Cyberscience Center Tohoku University |
Date | 2010-01-22 |
Paper # | AI2009-21 |
Volume (vol) | vol.109 |
Number (no) | 386 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |