基于网络评论文本挖掘的技术预见方法研究

Technology Foresight Method Study Based on Text Mining of Web Comments

摘要:

在网络社区兴起的背景下，鉴于网络社区的海量评论数据中蕴含着大量专家用户群体智慧，本文提出基于网络评论文本挖掘的技术预见新型方法，以促进技术预见活动顺利实施并取得准确可信的最终结果。首先从多源数据中获得种子科技主题，并将其投放至开放网络社区，吸引专家用户进行充分讨论形成交互数据，经过数据爬取、清洗、存储等环节得到网络评论数据集，再利用情感分析、主题模型等方法对网络评论中蕴含的隐性知识进行显性化挖掘，并结合相关领域专家的研判，最终得到辅助技术预见决策的有价值信息。通过新型方法，可以使技术预见活动大幅降低成本、打破时空限制，便于大规模专家参与其中，并最大限度降低少数专家主观色彩浓厚的负面影响。

Abstract:

In the context of the rise of online communities, considering the massive comment data in online communities contains a large amount of collective intelligence, a new technology foresight method based on text mining of web comments is proposed, in order to promote the smooth implementation of technology foresight activities and obtain accurate and credible final results. Firstly, technology innovation topics are obtained from multi-source data and put into online communities, attracting experts and users to discuss to form interactive data, after data crawling, cleaning, storage and other procedures, web comment data sets can be obtained. Then, implicit knowledge contained in web comments is explicitly mined by sentiment analysis and topic model. And finally, valuable information is obtained after combining the judgment of experts in related fields. Through the new method, technology foresight activities can greatly save costs, break the time and space constraints, facilitate a large number of experts to participate, and minimize the negative impact of the subjective influence of a few experts.