常用个性化推荐算法的技术研究(论文12500字)
摘要： 随着因特网的出现和普及，它给我们带来了海量的信息，但是同样由于网络的迅速发展而带来的网上信息量大幅增长，让我们很难从海量信息中找到对自己有用的信息，目前针对此类现象的一般解决方法就是以搜索引擎为代表的信息检索系统，比如百度，搜狗等，而信息及其传播是多样化的，如今用户对信息内容以及信息数量的需求是具有多元化的并且极具个性化的，因此处理信息过量问题的另一个非常立竿见影的办法就是使用推荐系统。目前我们所研究的推荐算法主要包括基于内容的推荐、基于知识的推荐、关联规则的推荐、协同过滤推荐、基于社会网络分析法的推荐、基于网络结构的推荐和混合推荐。推荐系统的体系结构还有性能评价也是推荐系统的一个研究方向。
关键词：推荐算法；个性化推荐技术；社交网络；信息超载；
Commonly used technology research of the personalized recommendation algorithm
Abstract：With the emergence and popularity of the Internet, it brings us a vast amounts of information, but also because of the rapid development of the network of online information, it's hard to found in the mass information about their useful information, for this kind of phenomenon, the general solution is represented by the search engine information retrieval system, such as baidu, sogou, etc., and information and its transmission is diverse, now the requirements of the users of information content and information quantity is diversified and personalized, so deal with information overload problem of another very quick way is to use the recommendation system. Now we study the recommendation algorithm mainly includes recommendations based on content, based on the recommendations from the knowledge, the recommendation of association rules, collaborative filtering recommendation, recommendations based on social network analysis method, based on the recommendations from the network structure and mixed is recommended. Recommendation system architecture and performance evaluation is also a research direction of the recommendation system.
Key words：Recommendation algorithm; personalized recommendation system; social networking; information overload;
目 录
摘要及关键字………………………………………………………3
推荐系统的概念及研究背景………………………………………5
推荐系统的研究方向………………………………………………6
个性化推荐算法技术研究…………………………………………7
3.1 基于内容的推荐………………………………………………………………………7
3.2 协同过滤推荐…………………………………………………………………………8
3.2.1基于用户的协同过滤推荐……………………………………………………9
3.2.2基于物品的最近邻推荐………………………………………………………11
3.3 基于社交网络的推荐…………………………………………………………………11
3.3.1基于邻域的社交网络推荐……………………………………………………11
3.3.2基于网络结构的社交网络推荐………………………………………………12
3.4 基于知识的推荐………………………………………………………………………13
3.5 组合推荐………………………………………………………………………………13
3.5.1推荐结果的混合………………………………………………………………14
3.5.2推荐算法的混合………………………………………………………………14
3.5.3相似度的计算…………………………………………………………………14
3.6 部分推荐技术的比较…………………………………………………………………15
对于组合推荐算法的想法…………………………………………16
结论…………………………………………………………………17
参考文献……………………………………………………………19
致谢…………………………………………………………………20
|
