z-logo
Premium
A multigranular linguistic content‐based recommendation model
Author(s) -
Martínez Luis,
Pérez Luis G.,
Barranco Manuel
Publication year - 2007
Publication title -
international journal of intelligent systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.291
H-Index - 87
eISSN - 1098-111X
pISSN - 0884-8173
DOI - 10.1002/int.20207
Subject(s) - computer science , context (archaeology) , forcing (mathematics) , product (mathematics) , term (time) , perception , service (business) , content (measure theory) , recommender system , information retrieval , natural language processing , artificial intelligence , linguistics , mathematics , psychology , marketing , philosophy , mathematical analysis , paleontology , physics , geometry , quantum mechanics , neuroscience , business , biology
Abstract Recommendation systems are a clear example of an e‐service that helps the users to find the most suitable products they are looking for, according to their preferences, among a vast quantity of information. These preferences are usually related to human perceptions because the customers express their needs, taste, and so forth to find a suitable product. The perceptions are better modeled by means of linguistic information due to the uncertainty involved in this type of information. In this article, we propose a content‐based recommendation model that will offer a more flexible context to improve the final recommendations where the preferences provided by the sources will be modeled by means of linguistic variables assessed in different linguistic term sets. The proposal consists of offering a multigranular linguistic context for expressing the preferences instead of forcing users to use a unique scale. Then the content‐based recommendation model will look for the most suitable product(s), comparing them with the customer(s) information according to its resemblance. © 2007 Wiley Periodicals, Inc. Int J Int Syst 22: 419–434, 2007.

This content is not available in your region!

Continue researching here.

Having issues? You can contact us here