报告题目:Structural Equation Modeling with Text Data
报告人:Zhiyong Zhang, University of Notre Dame
报告时间:2024年5月17日(星期五) 上午9:50-11:30
报告地点:教2-327
主办单位:350VIP浦京集团350VIP浦京集团、随机数学研究中心、科技处
邀请人:温勇
报告内容:Text data are widely collected in research and can come from many different sources. However, text data are largely under-analyzed in social, behavioral and education research. In this study, we present a general model that can combine structural equation models with text data. A two-stage method will be used to first extract the information from the text data through different methods such as sentiment analysis and text encoders and then the information is used in SEM. I will show several examples to illustrate the application of such as model. The method is implemented in an R package BigSEM and a related web app.
报告人简介:
张志勇于2008年获得弗吉尼亚大学量化心理学博士学位。他目前是美国圣母大学心理学系的教授。他的研究重点是在心理学、教育和健康研究领域开发新的方法和软件进行实际数据分析。他最近的研究重点是社交网络分析和文本挖掘。他的研究得到了美国教育研究所和国家科学基金会的资助。他是Society of Multivariate Experimental Psychology (SMEP)的会员,也是American Psychological Association (APA)的会士。他是期刊the Journal of Behavioral Data Science的主编,以及Multivariate Behavioral Research的副主编。他曾获得SMEP Early Career Award和APA Jacob Cohen Award。关于他的更多信息可在https://bigdatalab.nd.edu网站查阅。