Result: BioPyAssistant: a chatbot to support learning of Python programming

Title:
BioPyAssistant: a chatbot to support learning of Python programming
Authors:
Bottani, Samuel, Janiszek, David, Poulain, Pierre
Contributors:
Université Paris Cité (UPCité), Laboratoire de biochimie théorique Paris (LBT (UPR_9080)), Institut de biologie physico-chimique (IBPC (FR_550)), Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Institut de Chimie - CNRS Chimie (INC-CNRS)-Centre National de la Recherche Scientifique (CNRS)-Université Paris Cité (UPCité), Financement IDEX Université Paris Cité 2024-2025. Projet d'innovation pédagogique LLM@UPCité., Université Paris Cité
Source:
https://hal.science/hal-05340526 ; Université Paris Cité. 2025.
Publisher Information:
CCSD
Publication Year:
2025
Subject Terms:
Chatbot, Artficial Intelligence, Python, Retrieval-Augmented Generation RAG, Large language Models LLMs, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], [SDV.BIBS]Life Sciences [q-bio]/Quantitative Methods [q-bio.QM]
Document Type:
Report report
Language:
English
Availability:
https://hal.science/hal-05340526
https://hal.science/hal-05340526v2/document
https://hal.science/hal-05340526v2/file/article_biopyassistant_Guild.pdf
Rights:
http://creativecommons.org/licenses/by/ ; info:eu-repo/semantics/OpenAccess
Accession Number:
edsbas.3364A8DF
Database:
BASE

Further Information

The LLM@UPCité project at Université Paris Cité developed BioPyAssistant, a conversational chatbot designed to assist master’s students in learning Python programming. Created by Essmay Touami during her master’s internship, the assistant uses a Retrieval-Augmented Generation (RAG) approach based on existing course materials available online and in PDF form. Through a web interface, students can ask questions and receive concise, source-cited answers with direct links to relevant course sections.Between October and December 2024, BioPyAssistant was tested by 70 bioinformatics master’s students. Survey results showed that 66% used it to clarify course content, 27% explored topics beyond the course, 71% followed the provided course links, and 100% found it suitable for beginners. Overall, the chatbot improved students’ understanding of Python concepts and engagement with the learning material.