AI-Powered Learning Assistant with Advisory

DOI

10.2991/978-94-6239-654-8_46How to use a DOI?

Keywords

Retrieval-Augmented Generation; Semantic Retrieval; Large Language Model; Hybrid Retrieval; Vector Embeddings

Abstract

The Student AI Chatbot is at the forefront of issues which today’s students face in the digital age we see an exponential growth of digital educational resources which in turn overloads the students’ access points, which in turn produces fragmented information and inefficient study practices. Also we see that traditional tools like keyword based search engines or static knowledge bases fall short in giving out precise, detailed answers to complex academic questions which in turn causes cognitive overload, waste of time and in the end reduced learning performance. To that end this project has put forth a very advanced Retrieval-Augmented Generation (RAG) web app which we have developed on a Streamlit platform, which in turn is meant to present to students very accurate, timely and context aware answers to their questions which in turn we are putting out there in many different academic fields. By the use of AWS Bedrock’s Nova Micro model for what we see as light weight low latency inference and Cohere’s embed-english-v3 for high dimensional semantic embeddings we have put in place a very robust processing of multimodal inputs which include user uploaded PDFs and dynamically validated web data. We take in content, break it up into chunks and we also do a summary using PyMuPDF and abstractive LLM techniques which we then put in to MongoDB with geo spatial style indexing for very efficient hybrid retrieval which we do via BM25 lexical scoring and cosine similarity. We have a multi step relevance validation pipeline in place that which reduces the chance of out of context responses thus the responses are very much a product of the retrieved info. Also we are augmented by AWS CloudWatch for operational telemetry and we have put in an interactive Streamlit interface which gives real time feedback, contextual highlighting and also very easy session management which in turn we feel gives us a very scalable, secure and engaging platform. This solution not only streamlines access to reliable academic knowledge but also fosters a transformative learning experience, empowering students to navigate complex curricula with confidence and efficiency.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - S.D. Tharukesh
AU  - E. Nishaan
AU  - G. Thiraviaselvi
PY  - 2026
DA  - 2026/04/24
TI  - AI-Powered Learning Assistant with Advisory
BT  - Proceedings of the Global Conference on Sustainable Energy Systems, Smart Electronics and Intelligent Computing (GCSESEIC 2025)
PB  - Atlantis Press
SP  - 574
EP  - 584
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6239-654-8_46
DO  - 10.2991/978-94-6239-654-8_46
ID  - Tharukesh2026
ER  -