Behavioral interventions are an important strategy to address public health issues such as smoking cessation, diet, or physical activity, which are the underlying drivers of many chronic disease and health care costs. In particular, motivational interviewing (MI), a client-centered style of counseling, has shown consistent efficacy across multiple health behaviors. The effectiveness of MI in actual practice is highly dependent on adequate training, supervision, and quality feedback to MI counselors.
The goal of this project is to develop a system that can automatically understand and evaluate the characteristics of good counseling using Natural Language Processing, and learn to differentiate between good and bad counseling through data-driven analysis of MI sessions. The system will therefore be able to provide objective feedback about the counseling sessions and provide real-time quality feedback to assist counselor supervision and assure the quality of counseling interventions.
Published in the Proceedings of the 55th annual meeting of the Association for Computational Linguistics (ACL), 2017
Published in the Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics, 2017