The goal of this project was to understand how biologists search for information related to their research, and to generate design ideas that leveraged machine learning to help their research process. I interviewed nine biologists and created models of the interview data to understand the biggest design opportunities. I then brainstormed design concepts and validated them with interviewees and machine learning experts. Download Project Report (pdf).

Biologists' Research Process

Information Flow Problems in Biology Labs

Example Design Idea from Project

Instant review paper

When James sent Steve 40 papers to read, he asked his computer to generate an "instant review paper." The computer looked for all of the important findings in the papers and eliminates the any repeated information. It provided him a 20-page review document.

Paper tree

Steve typed in a protein name and biological process that the lab was studying and saw many papers in a tree diagram. The tree modeled the flow of ideas in the field, with each branch showing the growth of a new idea.

Finding new areas from paper discussion sections

Steve typed in a protein and biological process that he was interested in. From there, he was able to see all of the research questions that research papers had mentioned in their discussion sections.