基于自媒体的旅游需求信息分类及本体概念获取研究  

Research on the Classification of Travel Demand Information and the Acquisition of Ontology Concept Based on “We Media”

作　　者：李志义[1] 杨雄威[1] 王冕[1] 

机构地区：[1]华南师范大学经济与管理学院,广州510006

出　　处：《图书情报工作》2015年第23期106-114,共9页Library and Information Service

摘　　要：[目的 /意义]微信、微博等自媒体中隐含着大量的用户旅游消费需求的信息,将这些信息进行分类并依据分类结果构建需求本体,从而帮助企业分析和研究用户需求以获取巨大的商业价值。[方法/过程]利用SVM分类算法将微博信息分类并生成分类结果集,这些结果集中包含大量旅游相关概念的词汇,可以作为构建和扩展旅游需求本体的语料;然后通过调查各大旅游网站的类目确定旅游需求的核心概念,抽取分类结果中与旅游相关的概念。[结果 /结论]利用抽取结果匹配核心概念,生成扩展后的本体,使用HOZO本体编辑工具进行修改和完善,并呈现部分旅游需求本体。从实验结果看,本文所提方法能较为准确地对包含旅游需求的文本进行分类。[Purpose / significance]The ＂We Media＂such as We Chat and micro-blog,implies a large demand of tourism consumption information. The information is classified and the classification results are used to construct requirement ontology,which could help enterprises to analyze user needs to obtain huge commercial value. [Method / process]The SVM classification algorithm is used to generate the microblog information classification results,which include a large number of tourism related concepts of vocabulary,as the construction and expansion of the corpus of the ontology of tourism demand; and then investigates large travel website categories to determine the core concept of tourism demand.[Result/conclusion]The extraction results are used to match the core concept and generate extended ontology. Use HOZO ontology editing tools to modify and improve the present part of tourism demand ontology. From the experimental results,this method can be used to more accurately classify the texts which include tourism needs.

关 键 词：自媒体 旅游需求 本体分类 概念获取 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]