|
Title:
|
MODELS FOR MANAGING INCOMPLETE INFORMATION IN RECOMMENDER SYSTEMS |
|
Author(s):
|
L. Martíne , L.g. Pérez , M. Barrranco |
|
ISBN:
|
972-98947-8-7 |
|
Editors:
|
Nitya Karmakar and Pedro Isaías |
|
Year:
|
2004 |
|
Edition:
|
Single |
|
Keywords:
|
Recommender systems, preference modeling, uncertainty, accuracy. |
|
Type:
|
Oral Presentation - 20 minutes |
|
First Page:
|
307 |
|
Last Page:
|
312 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
Recommender Systems have recently emerged to assist network users in their search processes, due to the fact, these users must deal with a vast quantity of information that is stored in huge data bases of different e-shops, e-libraries, etc. The Recommender Systems help them in their search by means of recommendations that arise from information provided by different sources as the proper user, experts, other users, etc. Most of the current Recommender Systems force their users to provide the information in an only way, usually a numerical scale. Nevertheless, the information provided by the different sources get to use incomplete, vague and imprecise because it is related to their own perceptions, tastes and preferences. In other research areas as decision analysis, planning, scheduling, etc., this type of information has been successfully managed allowing the sources of information to express their preferences by means of different representation models and preference structures. In this contribution we shall review these models and structures together several resolution processes dealing with them in order to propose their use in the Recommender Systems to improve their accuracy and success in the recommendations. |
|
|
|
|
|
|
