基于WAE主题模型和社会网络的农产品在线交互研究——以“沙糖桔吧”为例

Online Interaction of Agricultural Products Based on WAE Topic Model and Social Network:A Case Study of “Sugar Orange Bar”

关键词:  农产品  交互行为  Wasserstein自动编码器  主题分析  社会网络分析

Key words:  agricultural products  interactive behaviors  Wasserstein Auto-Encoder (WAC)  topic analysis  social network analysis

摘要:

为了更好地利用农业信息资源，本研究将百度贴吧“沙糖桔吧”中的数据作为研究对象，对无序化的农产品在线交互信息进行数据挖掘，引入大数据技术促进农产品信息传播。首先，构建基于Wasserstein的自动编码器(Wasserstein Auto-Encoder,WAE)和高斯先验的主题模型(WAE-G)，将其应用于交互信息的主题识别和提取；然后，利用社会网络分析方法对交互用户进行社区聚类，探究聚类主题及网络中的关键用户。实证分析结果验证了该主题模型的有效性，所提取的主题信息能较好地反映用户的信息需求、市场供需现状以及该产业发展的瓶颈。同时，社会网络分析结果发现了不同社区用户交流的主题差异，以及关键用户在信息传播中的作用。

Abstract:

To make better use of agricultural information resources,this paper takes the data of “Sugar Orange Bar” on Baidu Post Bar as the research object to conduct data mining on the disordered online interactive information of agricultural products,so as to promote the application of big data technology in the information dissemination of agricultural products.Firstly,based on Wasserstein Auto-Encoder (WAE) and Gaussian prior,the topic model (WAE-G) is constructed and applied to the topic recognition and extraction of interactive information.Then,social network analysis is employed to cluster interactive users and explore the topics between clusters and the key users of the network.The results of empirical analysis verify the validity of the model,and the extracted topic information reflects the information demand of users,the current situation of market supply and demand,and the bottleneck in the development of the industry.At the same time,the results of social network analysis reveal the thematic differences of user communication in different communities and the role of key users in the dissemination of information.