Probabilistic Topic Modeling of User Search to Identify Tends in Research

• In Event: 63.084 - MTCC Roundtable Session 25
  In Roundtable Session: 63.084-7 - Advances in Learning Environment Studies

Abstract

The continued improvement in digital library service infrastructure enables collection of various access data. The large amount of accumulated data presents new opportunities for understanding modern library services. However, at the same time, the unstructured nature of the data poses challenges in analysis. In this paper, we introduce an innovative application of topic modeling of research literature extracted from library access data. The underlying topics of these literature coincides with the general interests one would expect from a Graduate School of Education. In addition, the results suggest little variance in topics of the papers published in different years.

Author

• Yi Chen, Columbia University