Research
Automatic evaluation of call center response quality
Overview
I research methods that automatically evaluate response quality from operators' response transcripts, using natural language processing and large language models.
Research fields
- Natural language processing (NLP)
- Practical applications of large language models (LLMs)
Background
When people evaluate response quality by hand, it takes time and money, and the standard can vary from one evaluator to another.
Purpose
To reduce the cost of manual evaluation and make the evaluation standard more consistent.
Technology and methods
| Item | Details |
|---|---|
| Data | Operators’ response transcripts |
| Methods | Natural language processing and large language models |
| Evaluation | Automatic evaluation of response quality |
Publications and results
These will be added here once they are confirmed for publication.