ECE 1786: Creative Applications of Natural Language Processing

Call for Projects

A key part of this course is the project, which is done in groups of 2. The students are trained in neural-network-based Natural Language Processing, through prior work in Machine Learning and the four assignments in this course.

If you have a specific project that requires anguage processing or language generation it could be an appropriate course project for one of our groups. If your project is taken up by a group in the course, you will have the opportunity to have your ideas tested, in the context of a for-credit course.

To do so, reach out to the instruction - Jonathan Rose (Jonathan.Rose@ece.utoronto.ca) with the following information:

A few paragraphs describing the project - its specific goal, and some motivation. (Two examples: (i)“I’d like to search transcripts of a therapy conversation for specific types of language; This part of a project on looking for positive and negative indications of mental health, for use in treatment tracking.”) (ii) “I’d like to automatically generate language that is similar to that produced by Frances Burney, to see if that is a style that can be imitated.”
An example of the language that would be the basis of the project - the “data” as it we would call it.
An understanding that you are able to share you data with students in the class. You can make non-disclosure of the data a condition of its use, but that may limit the uptake of the work.
Your contact information.
A commitment to spend 2-3 hours meeting with the team to describe the project, and availability after that to answer questions.