About

Large Language Models for Qualitative Analysis.

Paper Code Demo About

origin
Starting from two questions
2020 · I had two questions regarding qualitative analysis when I was first exposed to qualitative analysis: 1) Subjectivity in Interpretation: Qualitative analysis is inherently subjective. I observed that individuals with different backgrounds can have varying understandings or interpretations of the same data. This diversity in perspective was starkly different from the more objective nature of technological solutions; 2) Time Consumption for Analysis: The process requires a significant time investment, often spanning months, to analyze relatively small datasets. This seemed particularly inefficient and unreasonable in an era where machines and AI can automate numerous tasks. This contrast between manual analysis and technological efficiency was particularly striking for me at that time.
found
A new discovery
2021.08 · Found that the AI-assisted Collaborative Qualitative Analysis is important but has yet to be explored in academia.
2021.12 · Finished CoAIcoder prototype development
begin
The first step: CoAIcoder
2022.04 · Our study showed that while using a shared AI model as a mediator among coders could improve CQA efficiency and foster agreement more quickly in the early coding stage, it might affect the final code diversity. This made us submit the CoAIcoder manuscript to the TOCHI journal.
2022.09 · CollabCoder idea was proposed.
2023.04 · CollabCoder paper was on arXiv.
2023.06 · CoAIcoder paper was accepted to the TOCHI journal.
2023.10 · CollabCoder demo was displayed at CSCW2023.
2024.01 · CollabCoder full paper was accepted to CHI2024.
now
Current progress
2024.03 · CollabCoder's arXiv version gets 23 citations