基于RSS的中国林业网个性化信息服务研究

发布时间：2018-09-05 10:16

【摘要】：随着信息推送技术的不断发展,RSS服务作为信息推送服务的代表,不断的得到改进与发展,被广泛的应用于各个领域中。中国林业网作为中国林业政务第一门户网站,采用了“一站式”的建站方式,网站结构复杂,网站内信息量巨大。面对庞大而复杂的站点结构与海量的信息,用户在网站中主动查找符合自身需求的信息是困难的。因此,将RSS服务引入是有必要的。然而,传统RSS服务存在个性化程度低的问题,用户仅能依据服务商给定的信息频道去订阅信息,而无法根据自身兴趣创建个性化信息聚合。针对以上问题,本文首先将RSS服务引入中国林业网,构建中国林业网RSS个性化种子门户网站与RSS在线阅读器,为林业网用户提供基于推送技术的个性化信息服务。其次,将基于信息检索的RSS信息再聚合方法引入RSS服务中,实现基于个性化用户频道的信息再聚合。在RSS信息再聚合服务中,采用基于概念语义树扩展的检索方式,构建领域知识语义树与同义词表,通过引入扩展权重的特征成分单词权重计算与基于时间相关度的语料相似度计算,以期提高检索的准确率与召回率,完成高质量的RSS信息再聚合。最后,匹配RSS种子门户网站构建RSS在线阅读器。实验结果证明,本文的方法能够有效提高林业网信息检索准确率与召回率,能够为林业网用户提供高质量RSS信息再聚合服务。
[Abstract]:With the development of information push technology, RSS service, as the representative of information push service, has been continuously improved and developed, and has been widely used in various fields. As the first portal of forestry government in China, China Forestry Network adopts the "one-stop" mode of construction. The structure of the website is complex and the amount of information in the website is huge. In the face of huge and complex site structure and huge amount of information, it is difficult for users to actively find information that meets their needs in the website. Therefore, it is necessary to introduce RSS services. However, the traditional RSS service has the problem of low degree of personalization. Users can only subscribe to the information according to the information channel given by the service provider, but can not create personalized information aggregation according to their own interests. In order to solve the above problems, this paper first introduces RSS service into China Forestry Network, constructs RSS personalized seed portal and RSS online reader, and provides personalized information service based on push technology for forestry network users. Secondly, the method of RSS information reaggregation based on information retrieval is introduced into RSS service to realize the information reaggregation based on personalized user channel. In the RSS information reaggregation service, the domain knowledge semantic tree and the synonym table are constructed by using the retrieval method based on the extension of the conceptual semantic tree. In order to improve the accuracy and recall rate of retrieval, the RSS information can be reaggregated with high quality by introducing the extended weight calculation of the word weight of the feature components and the calculation of the similarity of the corpus based on the temporal correlation. Finally, match RSS seed portal to build RSS online reader. The experimental results show that this method can effectively improve the retrieval accuracy and recall rate of forestry network information, and can provide high quality RSS information reaggregation service for forestry network users.
【学位授予单位】：北京林业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TP391.3;F326.2

【相似文献】