|
ИСТИНА |
Войти в систему Регистрация |
ИСТИНА ЦЭМИ РАН |
||
Clustering of user profiles based on real implicit data in e-commerce recommender systemsстатья
- Авторы: Philippov S.A., Zakharov V.N., Stupnikov S.A., Kovalev D.Y.
- Сборник: Selected Papers of the XVII International Conference on Data Analytics and Management in Data Intensive Domains (DAMDID/RCDL 2016)
- Серия: CEUR Workshop Proceedings (CEUR-WS.org)
- Том: 1752
- Год издания: 2016
- Место издания: CEUR-WS.org
- Первая страница: 98
- Последняя страница: 103
- Аннотация: This work is devoted to description of key tasks in the context of building the online store information systems. The main objective is to identify user preferences and their formalization through the formation of the users’ behavioral profiles followed by the identification of user groups with similar characteristics. The main source of information about user preferences is implicitly an array of data collected about their actions when navigating through the pages of online shopping. The authors present an approach to building a recommendation system based on collaborative filtering problem solving using cluster analysis techniques to identify groups of users with similar preferences. Advantages of the solution are demonstrated on the example of test data set obtained from the current online store Thaisoap. A unique identifier of the project supported by the Ministry of education and science of the RF is RFMEFI60414X0139.
- Добавил в систему: Ступников Сергей Александрович