Exploring Deep Learning Models for Understanding Consumer Language

With the ultimate goal of enhancing Nexxt Intelligence’s market research SaaS platform, inca, this project will create a robust, scalable algorithm for clustering consumer utterances into groups which are useful to ad-hoc market research objectives, and interpretable via natural language descriptions to market researchers. Due to the multifaceted and nuanced nature of consumer opinions and feedback, this algorithm will leverage the latest advances in natural language processing for the purpose of representing semantics of complex sentences and phrases, detecting brand and product names even of hypothetical or proposed products, as well as extending work in human emotion modeling. The resulting algorithm will be applied to several downstream tasks, including a human-in-the-loop utterance analysis system and to a conversational framework for facilitating dynamic information-seeking dialogues.

Faculty Supervisor:

Annie Lee

Student:

Partner:

Nexxt Intelligence

Discipline:

Computer science

Sector:

Professional, scientific and technical services

University:

University of Toronto

Program: