“紫金主张”海外专家学者讲座预告

阅读: 10


讲座名称:Exploring Wikipedia-enhanced Concept Embeddings for   Education Content Representation

讲座时间: 627日(周四)下午3:00-5:00

讲座地点: 经管院106

讲座简介:Concepts are fundamental units which are used to summarize a learning content, tag a course, define prerequisites and state learning objectives for a course. In traditional education systems, concepts are defined either through expert-based manual annotations or automatic text-based key-phrase extraction. Both approaches have their own advantages and limitations, which include expert-based annotations are reliable but time-consuming and expensive, whereas automatic key-phrase extraction is quick but suffer through the problem of noisy and non-standard concept extraction. To address these challenges, we introduce one approach for concept embeddings, which can utilize domain specific educational content and external knowledge graph resources to achieve robust and effective concept representation. We conducted evaluations of the proposed models on two fundamental tasks in education system: education content linking and student modeling. The results demonstrate that our proposed concept embedding representations can perform better than the state-of-the-art automatic concept representation, and the representations can help address the challenge of bringing all concepts to one representation space.

主讲人简介:何大庆博士现为美国匹兹堡大学计算与信息学院(iSchool)教授,并担任iSchool图书馆与信息博士计划委员会主任。他担任了十余个研究项目的主持人或共同主持人,研究项目包括:美国国家科学基金会项目、美国国防部高级研究计划署资助项目、匹兹堡大学及其他机构资助的项目。何教授在国际公认的期刊与会议上发表论文120余篇,并担任信息检索及Web技术领域的二十多个主要国际会议程序委员会成员,同时担任SCI索引杂志Information Processing and ManagementInternet ResearchAslib Journal of Information Management的编委。其研究工作主要集中在:信息检索(单语言和多语言)、社交网络上的信息获取、自适应Web系统与用户建模、交互检索界面设计、Web日志挖掘与分析和研究数据管理。


评论功能已关闭。