以豆瓣网和中国国家图书馆为案例的网上书目推荐系统研究  

The research of online bibliographic recommendation:a case study on Douban.Com and National Library of China

作　　者：范旭[1] 

机构地区：[1]中山大学资讯管理系

出　　处：《图书馆学研究》2008年第8期44-48,共5页Research on Library Science

摘　　要：现有网上书目推荐系统根据应用服务理念的不同可以分为Web2.0型和传统型,根据其使用的过滤技术可以分为非个性化推荐、基于产品属性的推荐、基于相关用户的推荐和基于用户知识结构的推荐。书目推荐的开展应当遵循针对性、可行性、时效性这三个原则。本文通过对豆瓣网和中国国家图书馆的网上书目推荐服务的调查,从有用性、易用性、满意度、可找到性、普及性和可靠性六个方面对它们进行了比较,概括了传统型网上书目推荐系统的优势和改进意见。According to the service value, online bibliographic recommendation system is divided into two types： web 2.0 type and traditional type. On the base of filter technology, there are five types of recommendation system： Non- Personalized Recommendation, Attributed- Based Recommendation, People -to-People Correlation and Knowledge-Based Recommendation. Bibliographic recommendation service should follow three principles： accuracy, feasibility and timeliness. Through the case study of Douban. Corn and National Library of China, this paper compares them from six aspects： useful, usable, desirable, findable, accessible and credible. The advantages of traditional type online bibliographic recommendation systems as well as the opinions on its improvement are proposed.

关 键 词：书目推荐 图书馆导读 豆瓣网 中国国家图书馆 

分 类 号：TP319[自动化与计算机技术—计算机软件与理论]