IBM Watson - Cold Case Solutions

By Sijie Yang | Clinton McKay | Deepak Bhagchandani

My primary contribution in this project was coming up with concepts and creating every single piece of user interface below with high fidelity. We were tasked to create an app or service that leverages Watson’s cognitive computing capabilities and services in the cloud within one week. Requirements included a clearly defined problem within a vertical of our choosing, utilizing Watson in a meaningful way without over-stretching the capability, delivering a solution, app, or service that has compelling human value and crafting a set of deliverables that clearly communicate our process as well as the final result.

Based on our concrete understanding established from the requirements and initial research, we quickly scoped this design challenge by fully embracing Watson's constraints - it takes time and money to process huge quantity of fact-based text files in order to answer critical questions for people. Our brainstorming session helped us target at solving cold cases which had the greatest potential of allowing us to take advantage of Watson's constraints. After looking into not only how the existing system works, but also how police departments deal with cold cases, we started sketching our ideas and crafting the scenarios.

The final concept emerged with usability testing and following iterations. Reaching out to local detectives and asking for feedback from IBM made us be able to refine it even further. Both desktop and mobile solutions were presented to the design team at IBM eventually.