UX Research  |  Evaluative Research |  Enterprise Design Thinking | B2B

AutoAI with IBM Watson Studio-- UX Research Project 

How to create a better automatic modeling experience for data scientists?

March 2020 - Now, Senior Design Researcher Co-op

 

Summary

AutoAI in Watson Studio is an auto machine learning tool and can automatically generate models based on different algorithms. It helps data scientists to choose the optimal model from the dozens of pipelines. It saves them much time to build models.

 

Objectives

  • In what areas of the UI do users struggle?

  • What design improvements are needed to better enable existing and potential new users?

  • How can we help the data scientists more easily learn to use new features and find the optimal model?

To comply with the non-disclosure agreement, I have omitted and obfuscated confidential information in this case study. All information in this case study is my own and does not necessarily reflect the views of IBM.

My role

As a UX Research Intern at Watson Studio, I keep tracking with AutoAI.

  • Authored research plan, script, and tasks.

  • Recruited enterprise users, and coordinated interview schedules.

  • Set up and facilitated all the testing sessions.

  • Synthesized raw data to do quantitative and qualitative analysis 

  • Delivered issues and insights to the design team and IBM Research team.

My team:

Yanbin Hao (Design Researcher intern)

Dillon Eversman (UX Designer)

Voranouth Supadulya (UI Designer)

Methods​

Heuristic Evaluation

Expert Review

Task Analysis

Moderated Usability Testing

Semi-Structure interview 

Cognitive Walkthrough

Concept Testing

Deliverable Results

Reseach Deck & Research Report

Research Process

Data Analysis

Data Sample

Deliverable Results

  • Research Deck

  • Research Report

Impacts

What I have learned:​