摘要
随着各类移动端应用与网页端应用技术的不断发展,各类推荐系统与人类生活联系逐渐变得更为密切;用户对于推荐系统的推荐效果要求日益提高,持续单一的推荐内容已经不能满足用户不断提高的要求,因此如何精准对接用户需求,解决数据稀疏问题并提供给用户更为精确的推荐效果都已经成为推荐问题中亟待解决的问题。同时各种推荐系统技术不断发展,文章对近年来各种基于MAB算法的推荐系统的研究动态和最新进展进行了综述,对其基本概念和算法的核心思想以及评价指标诸如点击率,运行时长以及累积遗憾等方面进行了分析比较,并对推荐系统技术的发展趋势和应用前景进行了预测。
With the continuous development of various mobile terminal applications and web application technologies,various recommender systems have gradually become more closely related to human l*ife.Users requirements for the recommendation effect of the recommendation system are increasing day by day,and the continuous single recommendation content can no longer meet the constantly improving requirements of users.Therefore,how to accurately connect the needs of users,solve the problem of data sparsity and provide users with more accurate recommendation effect has become an urgent problem to be solved in the recommendation problem.Various recommendation systems technology development at the same time,in this paper,the various recommendation system based on MAB algorithm in recent years,the research of the dynamic and the latest progress were summarized,the basic concepts and algorithms of core idea and the evaluation index such as clicks,running time and cumulative regret is analyzed and compared,and the recommendation system technology development trends and application prospect is forecasted.
作者
陈珂
Chen Ke(Nanjing University of Posts and Telecommunications,School of communication and information engineering,Nanjing,210003)
出处
《长江信息通信》
2021年第3期43-46,共4页
Changjiang Information & Communications
