Name
Abstract | ||
|---|---|---|
Due to the development of the Internet and smart technology, massive amounts of data with transaction records have been generated by online and offline environments. And the proliferation of items has made it difficult for customers to find the specific items they want to buy. In order to solve this problem, many companies have adopted recommender systems to provide personalization services. However, due to the explosive growth of data, they try to use only meaningful and essential data in order to reduce these costs. And, because recommender systems necessarily deal with personal and sensitive information, some customers are concerned that their private information may be exposed by them. Based on these concerns, in this study, we analyze the effects of the amount of information on the recommendation performance. We assume that a customer could choose to provide overall information or partial information. Using two data sets which are obtained by on-line and off-line environments, we evaluate the difference in the performance of customers who provided overall information and partial information. The experimental results indicated that the recommendation performance for customers who provided overall information generally shows higher accuracy but there are some differences between on-line and off-line environments. Therefore, our study can provide some insight to companies concerning the efficient utilization of data. |
Year | Venue | Field |
|---|---|---|
2016 | ICEC | Recommender system,Collaborative filtering,Computer science,Online and offline,Information sensitivity,Private information retrieval,Marketing,Personalization,Information filtering system,The Internet |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
9 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Hyun Sil Moon | 1 | 4 | 1.41 |
| Jung Hyun Yoon | 2 | 1 | 0.71 |
| Jae Kyeong Kim | 3 | 1011 | 52.32 |