报告题目：A Joint Model of Knowledge Topic-Structure Exploration for Online Content
报告人：唐洪婷 博士
时间：2018年9月29日 15:30-16:30
地点：管理学院第三会议室
报告内容摘要：
Researchers and practitioners have long realized the importance of innovation knowledge. However, it is a challenge task to identify valuable knowledge of interest from unstructured texts. In this study, we proposed an automatic joint model for simultaneously extracting knowledge topic and knowledge structure from online contents. By employing a variant of Latent Dirichlet Allocation (LDA) model, one important contribution of our work is the design of hotspot topic model that combine text attributes and user interests. Another major contribution of our research is the design of a novel framework for knowledge structure construction that underpinned by association analysis. We evaluate the quality of discovered information from both statistical fit and substantive fit. Based on our experimental results, we conclude that our proposed method exhibits a better performance compared with the contrast method, and could find more meaningful innovation topics. The managerial implication is that stakeholders can apply the proposed model to enhance their product or service decision making, particularly when market evolution trend may not be catchable.
【报告人简介】
唐洪婷，华南理工大学管理科学与工程系博士，新加坡国立大学信息系统与分析系联合培养博士，主要研究方向为知识管理、数据挖掘、社会网络、用户识别、区块链等。唐洪婷博士利用在线社区的用户生成内容在知识挖掘和用户识别相关领域进行了一系列具有创新性的研究，在SCI、CSSCI及EI检索期刊上以第一作者或通讯作者的身份发表学术论文7篇。同时作为主要项目完成人，参与国家自然科学基金项目“基于超网络的大众协同创新社区用户知识主体挖掘方法研究”，“移动社交应用成瘾性使用判别、行为特征挖掘与智能识别”，“移动社交网络成瘾性使用的驱动因素与负面影响研究”等。
欢迎有兴趣的教师，全体博士生、硕士生参加。
学院高层次人才引进办公室
2018-09-25
