This is my capstone project at UT-Austin MSIS 2023. The open-ended question is a widely used method in human-subject surveys. Open-ended questions provide opportunities for discovering human subjects’ spontaneous responses to the questions. However, it is labor-intensive and time- consuming to analyze the responses. In light of the challenge, I aim to leverage NLP methods to create a tool for assisting the analysis of open-ended questions.
I extracted the codes using PyABSA. The sample data is also aquire from the repo.
I implemented the classifier leveraging GPT-neo released by EleutherAI. The model can be accessed on Hugging Face
- Packages
pip install -r requirement.txt
- Codes Extraction
python main.py --file_path <path to responses file> --extraction
The codes will be saved as codes.csv. You can edit the codes before you do classification.
- Classification
python main.py --file_path <path to responses file> --classification
The results will saved as json files.
You can also run python main.py --file_path <path to responses file> --extraction --classification. However, you will not be able to edit the codes.