|
Accession number;07A0065918
|
| Title;A Book Recommendation Model for Ubiquitous Computing |
| Author;
NAKADA TOYOHISA
(Japan Advanced Inst. Sci. and Technol., Hokuriku)
KANAI HIDEAKI
(Japan Advanced Inst. Sci. and Technol., Hokuriku)
KUNIFUJI SUSUMU
(Japan Advanced Inst. Sci. and Technol., Hokuriku)
|
Journal Title;IPSJ Transactions on Database
|
Journal Code:Z0778A
|
ISSN:0387-5806
|
|
VOL.48;NO.1;PAGE.148-162(2007)
|
| Figure&Table&Reference;FIG.11, REF.19 |
| Pub. Country;Japan |
| Language;Japanese |
| Abstract;The advantage of a real bookstore over an online bookstore is that user can touch and read books. In general, such behavior helps user to increase the magnitude of an interest of a book and enhance purchasing it. In order for user to touch and read many books, we focus a book recommender system for real bookstore and propose a recommendation model for the system. The recommendation model used on online bookstores is often deterministic during execution because the one of the main purpose of the recommendation model is to reduce computational complexity in order to handle huge amount of the book information and buying histories by users. However, the model cannot handle dynamic behavior of users. Our proposed model, in contrast, can handle it using probabilistic model. In the article, we describe our proposed model and show the results from experiments that is performed in real bookstore simulation environment. (author abst.) |
|
|
|
Related Articles;
|
