Perfil de usuários da Biblioteca Karl A. Boedecker
Carregando...
Arquivos
Data
2005-11-24
Orientador(res)
Métricas
Título da Revista
ISSN da Revista
Título de Volume
Resumo
Artificial Intelligence techniques can be applied to library circulation transactions to generate suggestions of items relevant to patrons and researchers. The Recommendation System presented in this report is based on consolidation and linkage of transaction records stored in a Circulation Data Mart, and on Basket Analysis, Cluster Analysis and Link Analysis techniques. An strategy of indirect cooperation has been adopted. In the proposed model, library items are consolidated in Significant Subjects and Theme Groups. Patrons, on the other hand, are clustered by Theme Group and segmented by their reading profile. Through consolidation, recommendation lists are generated for each Specialized Subgroup. Lists contain specialized, thematic and general suggestions. Virtual transactions are used to improve recommendations. The Recommendation System developed in this project can be used not only in libraries but also in virtual bookshops; it can be easily adapted to practically any kind of e-business enterprise.