Personalized News Recommendation: A Review and an Experimental Investigation
Personalized News Recommendation: A Review and an Experimental Investigation
复制标题
DOI：
10.1007/s11390-011-0175-2
复制
发表时间：
2011-09-01
影响因子：
1.9
通讯作者：
Li, Tao
中科院分区：
文献类型：
--
作者：
Li, Lei;Wang, Ding-Ding;Li, Tao
网络新闻作为一种新的新闻稿形式在互联网上兴起。由于其方便性和时效性，越来越多的人更喜欢在线阅读新闻而不是阅读纸质新闻稿。然而，海量的新闻事件可能会以每小时数百甚至数千的速度发布。一个具有挑战性的问题是如何从大量新发布的新闻稿中高效地选择特定的新闻文章推荐给个人读者，其中所选的新闻文章应尽可能符合读者的阅读偏好。本期涉及个性化新闻推荐。近年来，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，个性化新闻推荐已成为一个有前途的研究方向。现有的个性化新闻推荐系统致力于利用用户信息和新闻内容信息来使其服务适应个体用户。人们已经提出了多种技术来解决个性化新闻推荐，包括基于内容的协同过滤系统以及这两者的混合版本。在本文中，我们对现有的个性化新闻推荐器进行了全面的调查。我们讨论了个性化新闻推荐问题背后的几个基本问题，并探索了性能改进的可能解决方案。此外，我们对从各个新闻网站获得的新闻文章集合进行了实证研究，并评估了不同因素对个性化新闻推荐的影响。我们希望我们的讨论和探索能为对个性化新闻推荐感兴趣的研究人员提供见解。
Online news articles, as a new format of press releases, have sprung up on the Internet. With its convenience and recency, more and more people prefer to read news online instead of reading the paper-format press releases. However, a gigantic amount of news events might be released at a rate of hundreds, even thousands per hour. A challenging problem is how to efficiently select specific news articles from a large corpus of newly-published press releases to recommend to individual readers, where the selected news items should match the reader's reading preference as much as possible. This issue refers to personalized news recommendation. Recently, personalized news recommendation has become a promising research direction as the Internet provides fast access to real-time information from multiple sources around the world. Existing personalized news recommendation systems strive to adapt their services to individual users by virtue of both user and news content information. A variety of techniques have been proposed to tackle personalized news recommendation, including content-based, collaborative filtering systems and hybrid versions of these two. In this paper, we provide a comprehensive investigation of existing personalized news recommenders. We discuss several essential issues underlying the problem of personalized news recommendation, and explore possible solutions for performance improvement. Further, we provide an empirical study on a collection of news articles obtained from various news websites, and evaluate the effect of different factors for personalized news recommendation. We hope our discussion and exploration would provide insights for researchers who are interested in personalized news recommendation.