API Usability of Machine Learning Libraries

API usability specifies how easy, efficient, error-preventing, and pleasant an API of a software library is from its users’ perspective. With machine learning (ML) techniques becoming increasingly powerful and pervasive, many non-programmers
and casual users (e.g. domain experts in medicine or geography) started to explore ML libraries. However, many find them challenging to use because of bad API design. This project aims to investigate the API of ML libraries through the lens of
user-centered design. The knowledge gathered will help developers of ML libraries improve their APIs and establish preliminary methodologies to evaluate API usability of ML libraries.

Faculty Supervisor:

Jinghui Cheng

Student:

Partner:

National Taiwan University

Discipline:

Engineering

Sector:

Technology; Information and Communications Technology

University:

Polytechnique Montréal

Program: