Meet the Watson – AI Computer and a Champion in Jeopardy! Quiz
26 May 2017
26 May, 2017

The Watson supercomputer was developed in the IBM DeepQA project, by a team of engineers led by principal researcher David Ferrucci. The computer was named after IBM's first CEO, industrialist Thomas J. Watson.

 

Its first goal was to replicate (or surpass) a human’s ability to answer questions, so Watson got access to 90 servers with a combined data storage of over 200 million pages of information, which it processes against six million predefined logic rules. The device and its data are stored in a space that is the size of an average room.

Watson Hardware 

Just to give you a better understanding of how powerful Watson is, we'll list some of its main hardware components.

 

 

Watson's key components include:
 

- 15 terabytes of RAM.

- 2,880 processor cores.

- 500 gigabytes of preprocessed information.

- SUSE Enterprise Linux Server 11, the fastest available Power7 processor operating system.

- Apache UIMA (Unstructured Information Management Architecture) infrastructure, frameworks and other elements needed for the analysis of unsorted data.

- Apache's Hadoop, a programming framework that enables the processing of large databases in the distributed computing environment.

- IBM'sDeepQA software, which is specially created for information retrieval that involves natural language processing and machine learning.

 

Also, Watson can process at a rate of 80 teraflops (trillion floating-point operations per second).

Watson Real-World Applications

Applications for the Watson's Artificial Intelligence are almost endless. The computer can perform data mining and sophisticated analytics on massive volumes of unstructured data; it can also support a search engine with capabilities far superior to any currently in existence. Last year, BakerHostetler, a century-old Ohio-based law firm, signed a contract with IBM for a legal expert system based on Watson to work with its 50-member bankruptcy team. The system can mine data from nearly a billion text documents, process the information and provide precise responses to complicated questions in less than three seconds. Natural language processing enables the system to translate legal documents into responses to the lawyers’ questions. Similar expert systems are already being used to transform the medical research.

 

To showcase Watson's abilities, IBM challenged two top-ranked players on Jeopardy! Quiz, and Watson was able to beat champions Ken Jennings and Brad Rutter in 2011. The Watson avatar sat between the two human contestants, to show he's physically present, while its considerable part sat on a different floor of the building. Of course, as the rules suggest, Watson had no Internet access.

 

Since then, the company has been preparing Watson for work in research, business, and medicine, aiming to help organizations get the answers to the questions faster and at a lower cost.

 

Currently, companies can choose between 28 application programming interfaces (APIs) which can be used to build Watson applications or integrate Watson’s capabilities within predeveloped systems of their own. The APIs can help analyze the tone of the text, script conversations, create a list of contextually related terms and classify natural language, and are entirely available from IBM’s cloud platform Bluemix

 

Applications that help companies make sense of huge volumes of written material could particularly benefit from using Watson.

 

For example, consultancy firm Deloitte is working with IBM’s team to offer a service that can gather big volumes of legal information that wouldn't be humanly possible, helping businesses save money on regulatory compliance.

 

“Some businesses might have 20,000 pages of regulations to sift through every month to keep on top of compliance,” says Surya Mukherjee, senior analyst at research firm Ovum. “To understand what is relevant to them, it takes an army of lawyers. The Watson application can parse the documents, and because it knows what to look for, flag up the relevant parts.” 

 

And what's more important, Watson learns from its errors. Of course, the system isn't perfect - there are false positives and false negatives, but with specialized algorithms and human feedback, the software learns from its mistakes over time.

Watson Brussels Summit 2017

The IBM hosts a Watson summit in Brussels on 30 & 31 May 2017. The event will incorporate workshops, round tables, demos, and presentations for IT, Finance and Security professionals.

 

Experts from all around the world will attend the summit and showcase how different companies and organizations put cognitive technology into practice. 

 

Below we posted the program and location of the summit. 

 

Program:

May 30 - Agenda IT and Innovation Track

May 30 - Agenda Security Track

May 31 - Agenda Finance Track

 

For a full program, please follow the official link.

 

Location: IBM, Bourgetlaan 42, 1130 Brussels
 

See you in Brussels! Don't forget to share your experience in the comments!

Ready to get Started?