Automating Textbook Requirements From Course Syllabi and Availability Tracking For Libraries

By Parsa Mallik

Introduction

My name is Parsa Mallik, and I am a Computer Science major. My senior capstone project focuses on creating a tool to aid college libraries in providing students with resources.

Abstract

This project focuses on developing a tool designed to automate the extraction and tracking of required and recommended readings from course syllabi at Earlham College. The tool utilizes zero-shot named entity recognition (NER) to parse syllabi and accurately identify textbook information, even when presented in varied and incomplete formats. Extracted data is stored in a database and further refined by web scraping to verify the availability of these textbooks in the Lilly Library’s catalog. The tool distinguishes between physical books and eBooks and updates the database accordingly. By automating this process, the tool will enable the Lilly Library to proactively ensure that essential textbooks are available to students, potentially reducing the financial burden on them. This project not only aims to benefit Earlham College but also holds the potential to be adapted for use in other educational institutions globally, promoting broader access to academic resources.