Using LLMs to Investigate Correlations of Conversational Follow-up Queries with User Satisfaction

This repository contains the LLM prompts for classifying conversation logs used in the paper "Using LLMs to Investigate Correlations of Conversational Follow-up Queries with User Satisfaction".

For Axis 1 (Purpose) classification, use Classifier_Axis1.py. For Axis 2 (Action) classifiaction, use Classifier_Axis2.py.