个性化医疗信息推荐系统的研究与实现

发布时间：2018-10-10 17:18

【摘要】：随着互联网时代的到来,网络上的信息呈现出指数增长的趋势。医疗信息资源作为海量网络信息中的一部分也呈现出了“爆炸式”的增长趋势。用户在海量网络信息中很难快速地找到自己所需要的有用信息,这就是在所谓的“信息爆炸”和“信息过载”的互联网时代所体现出来的弊端。微软亚洲研究院负责搜索的一名技术专家说：75%的内容通用搜索引擎搜索不出来,仅仅能够获取互联网中的一小部分信息。与此同时,通用搜索引擎往往返回给用户几十页甚至上百页的信息,但用户通常不会一页页的去查看是否是自己所需的信息,于是就导致了用户真正所需的信息可能出现在几十页甚至上百页之后而并没有被挖掘推荐出来。这就说明,通用搜索引擎虽然能够很轻松地帮助我们找到海量的信息,但是我们却很难从中找到自己真正想得到的信息。为了改善通用搜索引擎的弊端,本文研究并设计了一套面向医疗领域个性化的医疗信息推荐系统,该系统能够将用户所需信息及其相关的信息推荐给用户,能够很好的满足用户的需求。本文以数据挖掘和信息推荐算法为基础,研究设计并实现了一套专门用于医疗信息领域的个性化信息推荐系统。首先,本文详细讨论了个性化医疗信息推荐相关的关键技术,主要包括用户兴趣模型的构建和信息推荐基本算法,重点分析了几种信息推荐算法的优缺点,并最终设计了一种符合本系统设计所需的推荐算法。其次,详细阐述了医疗信息领域中个性化信息推荐系统的设计,从系统需求出发,构建系统的整体框架,设计了用户兴趣模型、中文分词模块、信息预处理模块、信息推荐模块以及个性化页面定制模块等。最后,实现了医疗领域中的个性化信息推荐系统,并对实验结果进行了分析。实验结果表明,在实验室环境条件下,本系统能够很好的将用户所需的医疗信息推荐出来,并且还能够推荐给用户一些相关的医疗信息。
[Abstract]:With the advent of the Internet era, the information on the network shows an exponential growth trend. As a part of massive network information, medical information resources also show an explosive growth trend. It is very difficult for users to find the useful information they need quickly in the mass network information, which is the drawback of the so-called "information explosion" and "information overload" in the Internet era. A search technology expert at Microsoft Research Asia said: 75 percent of all content search engines fail to search and can access only a fraction of the information on the Internet. At the same time, generic search engines often return dozens or even hundreds of pages of information to users, but users usually don't page by page to see if it's what they need. As a result, the information the user really needs may appear after dozens or even hundreds of pages without being mined and recommended. This shows that although the general search engine can easily help us find a huge amount of information, it is difficult for us to find the information we really want from it. In order to improve the disadvantages of general search engine, this paper studies and designs a personalized medical information recommendation system for medical field. The system can recommend the information needed by users and related information to users. Be able to meet the needs of users. Based on data mining and information recommendation algorithm, a personalized information recommendation system is designed and implemented in this paper. Firstly, this paper discusses the key technologies related to personalized medical information recommendation in detail, including the construction of user interest model and the basic algorithm of information recommendation, and analyzes the advantages and disadvantages of several information recommendation algorithms. Finally, a recommendation algorithm is designed to meet the need of the system design. Secondly, the design of personalized information recommendation system in the field of medical information is described in detail. Based on the system requirements, the overall framework of the system is constructed, and the user interest model, Chinese word segmentation module, information preprocessing module are designed. Information recommendation module and personalized page customization module. Finally, the personalized information recommendation system in the medical field is implemented, and the experimental results are analyzed. The experimental results show that under the condition of laboratory environment, the system can recommend the medical information needed by the user, and can also recommend some related medical information to the user.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：R319

【参考文献】