NLP-Based Adaptive Tutor: Shaping Personalized Learning Paths

Abstract

This research delves into the application of natural language processing (NLP) techniques to address complex questions within the context of computer science education. Utilizing connectivism as the theoretical framework, the study illustrates the efficacy of web scraping to gather substantial datasets from publicly available sources, applying these insights to inform educational practices. Furthermore, the research highlights the capability of NLP in extracting pertinent information from textual data, thereby supporting qualitative analysis. A practical example is provided, illustrating current job market trends for computer science students. The findings underscore the necessity to improve programming and testing skills within the curriculum. To support this, the paper presents a chatbot framework that employs LangChain and Streamlit, integrating various document types such as PDFs, DOCX, and TXT files. Powered by FAISS for vector-based document retrieval and Replicate’s Llama 2 for conversational AI, the system facilitates interactive question answering and document analysis, offering educators and researchers a valuable tool for efficiently gathering and analyzing knowledge.