移动环境下基于情境感知的数字图书馆个性化信息推荐研究

Context-aware Personalized Information Recommendation of Digital Library in the Mobile Environment

针对移动环境下读者个性化阅读需求具有高度情境敏感性的特点，面向数字图书馆领域提出一种基于情境感知的个性化信息推荐模型。首先，提出“情境熵”来度量读者对不同情境属性的情境感知度，并计算出每个情境属性在信息推荐中的相应权重值；在此基础上结合传统的协同过滤技术，提出一种基于情境感知的信息协同过滤推荐方法。最后，通过实验验证本文所提出方法具有有效性，能够较好地预测读者对情境属性的感知能力，为读者提供移动环境下个性化的阅读推荐服务。

Aiming at the high context-sensitive characteristic of readers＇ personalized reading needs in the mobile environment, this paper proposes a context-aware personalized recommendation model for the digital library field. Firstly, it proposes context entropy to measure readers＇ context perception on different context attributes, and compute the corresponding weights of each context attributes in the process of recommendation. Based on it, combined with the traditional collaborative filtering （CF） , a context-aware collaborative filtering recommendation approach is proposed. An experiment is conducted to verify the effectiveness of the proposed approach, which can predict readers＇ context pe,＇ception accurately and provide personalized reading recommendation services for readers in the mobile environment.