SCENE: a scalable two-stage personalized news recommendation system
SCENE: a scalable two-stage personalized news recommendation system
复制标题
DOI：
10.1145/2009916.2009937
复制
发表时间：
2011-07
期刊：
影响因子：
--
通讯作者：
Lei Li;Dingding Wang;Tao Li;Daniel Knox;B. Padmanabhan
中科院分区：
文献类型：
--
作者：
Lei Li;Dingding Wang;Tao Li;Daniel Knox;B. Padmanabhan
随着互联网提供了对来自世界各地的多个来源的实时信息的快速访问，推荐新闻文章已经成为一个有前途的研究方向。传统的新闻推荐系统努力通过用户和新闻内容信息来使其服务适应个体用户。然而，不同的新闻项目之间的潜在关系，和新的文章的特殊属性，如短的货架寿命和即时性的价值，使以前的方法效率低下。在本文中，我们提出了一种可扩展的两阶段个性化新闻推荐方法，该方法具有两级表示，该方法考虑了排他性特征（例如，新闻内容、访问模式、命名实体、流行度和新近度）。提出了一种基于用户兴趣的新闻选择原则框架，在推荐结果的新奇和多样性之间取得了较好的平衡。从各种新闻网站获得的新闻文章的集合上的广泛的实证实验证明了我们的方法的有效性和效率。
Recommending news articles has become a promising research direction as the Internet provides fast access to real-time information from multiple sources around the world. Traditional news recommendation systems strive to adapt their services to individual users by virtue of both user and news content information. However, the latent relationships among different news items, and the special properties of new articles, such as short shelf lives and value of immediacy, render the previous approaches inefficient. In this paper, we propose a scalable two-stage personalized news recommendation approach with a two-level representation, which considers the exclusive characteristics (e.g., news content, access patterns, named entities, popularity and recency) of news items when performing recommendation. Also, a principled framework for news selection based on the intrinsic property of user interest is presented, with a good balance between the novelty and diversity of the recommended result. Extensive empirical experiments on a collection of news articles obtained from various news websites demonstrate the efficacy and efficiency of our approach.