Automating Book-to-Curriculum Mapping Using Representation Learning

• In Event: Tools to Support Decision Making: Multi-Tiered System of Supports (MTSS), Machine Learning, and More?

Abstract

Reading is one of the most critical aspects of student learning. Historically, educators have selected curriculum-aligned literature manually, a process that is not only draining on already constrained time and resources but also overwhelming, given the nuances of curriculum standards and the large volume of new books published each year. In this paper, we present a methodology towards making this process of book-to-curriculum mapping more scalable and resourceful by leveraging the information contained in the back cover book descriptions. We provide a proof-of-concept analysis using the Common Core State Standards Mathematics and data from the California Department of Education's recommended book list. We demonstrate that book descriptions alone can be a sufficient source for effectively mapping entire books to curriculum standards.

Authors

• Yiwang Li, University of California - Riverside
  Presenting Author

• Yulie Park, University of California - Berkeley
  Presenting Author

• Xingyao Xiao, University of California - Berkeley
  Non-Presenting Author