| Proj No. | A3125-251 |
| Title | Socratic chatbot analysis – Part 1: NLP feature extraction and classification |
| Summary | Questioning is an important practice while learning. With an existing Socratic chatbot that has been developed and rolled out for a mathematics course in NTU, data has been collected on how students ask questions, and the responses generated by the chatbot to prompt them with more thought-provoking ideas. The conversation data along with other information such as timestamps and student background will be provided. This project will be split into 2 parts, with 1 student focusing on their specific task. The 2 students will finally work together and combine their respective segments to achieve the final product. Part 1 focuses on classifying the depth and quality of questions according to a taxonomy by extracting NLP features and using machine learning algorithms. The project will incorporate a mix of traditional feature extraction techniques as well as deep learning-based BERT variants compared against various large language models. This project will be in collaboration with NTU Applications of Teaching & Learning Analytics for Students (ATLAS) department under InsPIRE. |
| Supervisor | Dr S Supraja (Loc:School of Electrical and Electronic Engineering (EEE), Ext: +65 65148340) |
| Co-Supervisor | - |
| RI Co-Supervisor | - |
| Lab | Computer Engineering I (Loc: S2-B4c-15) |
| Single/Group: | Single |
| Area: | Digital Media Processing and Computer Engineering |
| ISP/RI/SMP/SCP?: |