There is an increasing interest in using machine learning(ML) across a variety of industries like finance, education and healthcare. Working on such ML product features, however, remains a point of friction for both non-technical and technical workers who without supporting infrastructure/technical knowledge struggle to fully use their skills in the context of ML. This thesis introduces novel approaches embodied in new systems to help facilitate workers better utilize the opportunities afforded by this rapidly evolving technology. At a high level, we rephrase problems encountered while ideating for ML-related features and utilizing scientific advancements as gaps in communication between the scientific community and the industry, and develop systems to help correct for this. Specifically, in the first project IdeaLens, we explore using real-world use-cases of past ML work as boundary objects while communicating technical abilities of ML work to help non-technical designers come up with new ideas within their domain. In the second project InToResearch we describe the design of a framework to spearhead an alternate ecosystem of ML research papers catered specifically to industry audience, and explore the use of TLDRs to help them navigate and find relevant information more efficiently.

66 pages

Thesis Committee:
Chinmay Kulkarni (Chair)
Sherry Tongshuan
Nikolas Martelaro

Srinivasan Seshan, Head, Computer Science Department
Martial Hebert, Dean, School of Computer Science

Return to: SCS Technical Report Collection
School of Computer Science

This page maintained by [email protected]