摘要
介绍推荐系统的推荐解释问题,分析提供推荐解释的多种方案,如基于协同的解释,基于内容的解释,基于知识和自然语言的解释和基于人口统计的解释,并阐述提供推荐解释的目的,如透明性、正确性、可信度等,为解决推荐系统的推荐解释问题提供一定的参考。
Introduces the recommendation explanation in recommendation systems, analyses several ways to provide recommendation explanation, such as collaboration based explanation, content based explanation, knowledge and natural language based explanation and population statistics based explanation, and expounds the purpose of providing recommendation explanation, such as transparency, validity and trustworthiness, which provides reference for solving recommendation explanation in recommendation systems.
基金
长春市科技计划项目(No.14JR002)
国家科技支撑计划项目(No.2013BAH07F00)
关键词
推荐系统
推荐解释
协同过滤
透明度
Recommendation System
Recommendation Explanation
Collaborative Filtering
Transparency
