Scientific Information Understanding

Date:  June  20 th, Tuesday,  2017

Start Time:  14:00

Venue: Room 720, Business  School Building

Speaker: Liu Xiaozhong  (Associate  professor, Department of Information and Library Science, School of Informatics  and Computing, Indiana University Bloomington)


Although  the scientific digital library is growing at a rapid pace, scholars/students  often find reading STEM literature daunting because of their limited knowledge  in the target domain and the challenging content of the readings. In this  research, I propose a new solution to address this problem: scientific  information understanding. Using novel, personalized, and communitized search  algorithms, a reader’s  emerging information needs can be projected onto a complex heterogeneous graph  (with complex graph schema), which hosts 32 types of relations, e.g., scientific  topic and math/algorithm formula evolution information. The proposed complex  graph navigation algorithm will optimize the personalized/communitized random  walk performance in a collaborative framework for cyberreading while helping  users access multi-modal Open Data Resources (ODR) for information  understanding. The goal of this proposed project is to develop a theoretical,  algorithmic, and systematic foundation for effective cyberlearning and  cyberreading by means of open-access online resources, novel text/graph, and  formula layout mining algorithms.