Book GPT: An Innovative PDF Querying Tool

Authors

  • D.SANTHAKUMAR Assistant Professor, Department of Computer Science &Engineering, SRM Institute of Science& Tech nology, Ramapuram Campus, Chennai, Tamil Nadu, India
  • L.SASIKALA Assistant Professor, Department of Computer Science &Engineering, SRM Institute of Science&Technology, Ramapuram Campus, Chennai, Tamil Nadu, India
  • A.BALAJEE Assistant Professor, Department of Computer Science &Engineering, Faculty of Engineering & Technology, Jain Deemed to be university, Bangalore, India

Keywords:

Natural Language Processing, k-Nearest Neighbor,OpenAI GPT-3

Abstract

Access to information that is timely and reliable is becoming increasingly important in today's fast- paced environment. A common format for transmitting information, PDFs frequently contain crucial data that must be retrieved and analyzed. Yet, it can be time-consuming anderror-prone to manually search through PDFs for the necessary data. In this paper, we describe a hybrid strategy that makes use of the capabilities of Natural Language Processing (NLP) methods to read and produce precise responses from PDFs. We employ k-Nearest Neighbor to locate the data points that are most pertinent to a particular query and Universal Sentence Encoder to convert sentences into fixed-length numerical vectors. We also incorporate the cutting-edge language model Open AIGPT-3 top roduce text that is similar to that of a human being for increased accuracy. By creating the interactive user interface, our tool will enable the users to either upload a pdf or to provide an appropriate URL to ask a query. Our method makes it possible to quickly and accurately extract answers from PDFs in real-time applications, such as those seen in the information retrieval, legal ,and health care sectors[1].

Downloads

Published

2023-11-10

How to Cite

D.SANTHAKUMAR, L.SASIKALA, & A.BALAJEE. (2023). Book GPT: An Innovative PDF Querying Tool. International Journal of Pharmacy Research & Technology (IJPRT), 14(1), 19–24. Retrieved from https://ijprt.org/index.php/pub/article/view/225

Issue

Section

Research Article