Automated Coding of Negotiation Transcripts

The Vanderbilt AI Negotiation Lab, in collaboration with NTR, has released two models that use Anthropic’s Claude LLM to code negotiation transcripts.  The Lab’s website is:  

https://www.ainegotiationlab-vanderbilt.com

Each of these two models apply specific coding schemes from prior projects.  The website explains the codes used and the model’s coding process, and reports how each model has been validated.  It also explains how to format your transcript data and how to submit your transcripts for analysis.

Model 1: https://www.ainegotiationlab-vanderbilt.com/model_aslani/

Model 2:   https://www.ainegotiationlab-vanderbilt.com/model_brett/

The history of how these models were developed can be found here:

History: https://www.ainegotiationlab-vanderbilt.com/static/assets/2407.21037v1.pdf

Costs:  Feel free to try a few transcripts.  If you plan to code large numbers of transcripts, we ask you to cover the fees we pay Anthropic to run the models.  Contact us if you need an estimate.   

If you have any questions, or if you would like us to build a model with your coding scheme (you must have a large set of human-coding transcripts to train the model), contact Ray Friedman at  Ray.Friedman@Vanderbilt.edu.