IBM
4 days
Dec, 2013
Watson is the name of the powerful supercomputer created by IBM. We decided to design a Watson application which can be effectively used to aid detectives in reopening and potentially solving cold cases, thereby saving the lives of innocent people.
BRIEF
Each team is tasked to create an app or service that leverages Watson's computing capabilities and services in the cloud. This app or service will serve as the flagship product for a company within any industry (e.g. healthcare, government, utilities) of your choosing. The CEOs of your companies have asked Watson integration consultants at IBM to vet the ideas, ensuring that each product is of high quality, provides lasting business value, and delivers a first-class user experience to the people who will be using your app or service (more specific details below). The broad ask: What could you create with Watson? How could you leverage cognitive computing to create a user experience above and beyond people's expectations? How will you delight people?
SOLUTION
We decided to leverage the power of Watson to help detectives in reopening cold cases. Cold cases are cases which are currently inactive due to the lack of a solid lead. Another characteristic of a cold case is that it has a finite set of information, which works well with Watson as it is only capable of learning from a finite set of textual information. Generally, a legal case involves a lot of information, which cannot all be perused by the detective together. Therefore, it could be possible that the detective did not make all the connections between the different data in all of the different pieces of information related to the case. Watson can step in at this point of time and try to find any clues of related data between the different pieces of information in a case, and thereby help the detective in solving the case. As a future strategy Watson can automatically fetch data from external databases (like credit card records, etc) so that it can work with a larger dataset and form stronger connections between the case facts. Round Two
PROCESS
Team members : CJ Yang + Clinton McKay + Deepak Bhagchandani)

When we initially started working on this project we first tried to narrow down our scope to a single domain that we were all interested in. After discussing the capabilities of Watson as described in the brief given to us, we realized that for Watson to work effectively on any domain it had to have the following characteristics:

  1. The corpus of information had to be finite

  2. The use case had to be justifiable to leverage the expensive use of the super computer

  3. The use case should work well with the time that Watson takes to learn new information ~ 6 months.


After brainstorming possible domain spaces that we would like to work with and which would work well with the constraints of Watson as listed above, we all agreed on using Watson to help detectives in solving cold cases, i.e cases which have become inactive due to the lack of solid evidence.
Research
Day 2 : Research After we decided to work with cold cases, we spent some time in trying to understand how cases go cold as well as how detectives work with cold cases currently. To do this, we watched documentaries related to cold cases as well as read information online. We also contacted cold case detectives in the Monroe County area but did not hear back from them on time.
Mock-ups
Day 3 & Day 4 : Mock-up Screens, Slide Deck After we had some background information about cold cases, we started drafting our user interface screens which would be interacted with by the detective. We also began creating our rationale and slide deck as well as our user scenario.
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
Projects