Scholarlink

In this paper, we present PDFQA, a PDF Question Answering System designed to extract information from PDF documents and provide accurate answers to user queries. The system leverages advancements in Natural Language Processing (NLP), utilizing frameworks such as Streamlit, PyPDF2, spaCy, and the Hugging Face Transformers library. PDFQA enables users to upload PDF files, pose questions related to the content of those files, and receive answers based on the extracted text. The system employs a combination of text extraction, semantic similarity analysis, and BERT-based question answering to deliver efficient and accurate responses. Through extensive testing and evaluation, we demonstrate the effectiveness and usability of PDFQA in navigating and extracting insights from PDF documents.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

PDFQA: A PDF Question Answering System

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Harihara Subramanian,
  • Shahawar Alim,
  • Jhevaan Reddy,
  • B. Prabha

摘要

In this paper, we present PDFQA, a PDF Question Answering System designed to extract information from PDF documents and provide accurate answers to user queries. The system leverages advancements in Natural Language Processing (NLP), utilizing frameworks such as Streamlit, PyPDF2, spaCy, and the Hugging Face Transformers library. PDFQA enables users to upload PDF files, pose questions related to the content of those files, and receive answers based on the extracted text. The system employs a combination of text extraction, semantic similarity analysis, and BERT-based question answering to deliver efficient and accurate responses. Through extensive testing and evaluation, we demonstrate the effectiveness and usability of PDFQA in navigating and extracting insights from PDF documents.