Watson is a computer that was created with the goal of being a question and answering machine that could process and organize large sets of data. The unique thing about Watson as compared to a regular search engine, Watson will break down the actual context of the question, try to understand it, and give a response as best it can with the information programmed in it's data base. This being different from a search engine which only searches for key words to try and direct you towards a site that can provide information similar to what you asked
.
Technological Components Behind Watson
Like many technological systems, Watson is comprised of many different components. There are 3 main components to Watson: it's software, its electrical components, and its data sources.
Electrical components
The main brain behind Watson is the POWER 7 processors. IBM has connected several (2,880 of them) together, comprising all of the servers needed to host Watson in order to give him the required mental capacity to process up to 1 million books per second.[1]
Software
IBM developed their own software for Watson to run they've coined as, "IBM's DeepQA Software." This system was written in various operating systems to provide the required coding.
Data Base
As we will get in to later in this article, Watson's knowledge base is unique to only what he is given. Depending on the course of work being done by Watson, he might be programmed with medical information as an example. He could have dictionaries, scholarly articles from the medical fields, diagnostics information, and just about any other pertinent information that could aid a nurse in determining the best treatment for their patient.
Watson's Thought Process
The fundamentals of Watson's operation starts with a question. Once asked, Watson will analyze the question, breaking it down, removing unimportant information. Watson then generates a hypothesis and begins to filter down his data base looking for possible outcomes to the question. Once done, Watson will then review his answers for supporting evidence and start assigning scores to his list of answers. As the list gets weeded out, Watson finally comes to his conclusion and gives a confidence score along with it as to assess how certain Watson is of his work. The following image breaks it down:
The Goal Behind Watson
The ultimate goal behind the creation of Watson was to create a technological system that could process data and give supported answers based on its data base. IBM decided the first test of Watson would be to see if they could create a system that could win a game against the most successful contestants of Jeopardy!
“The challenge is to build a system that, unlike systems before it, can rival the human mind’s ability to determine precise answers to natural language questions and to compute accurate confidences in the answers. This confidence processing ability is key. It greatly distinguishes the IBM approach from conventional search and is critical to implementing useful business applications of Question Answering.” Dr. David Ferucci, Principal Investigator DeepQA Watson Project.[2]
Watson successfully achieved this goal in 2011, so the goal line then moved to address the potential to evolve for commercial use.
Watson (IBM Computer)
Table of Contents
Watson is a computer that was created with the goal of being a question and answering machine that could process and organize large sets of data. The unique thing about Watson as compared to a regular search engine, Watson will break down the actual context of the question, try to understand it, and give a response as best it can with the information programmed in it's data base. This being different from a search engine which only searches for key words to try and direct you towards a site that can provide information similar to what you asked
.
Technological Components Behind Watson
Like many technological systems, Watson is comprised of many different components. There are 3 main components to Watson: it's software, its electrical components, and its data sources.
Electrical components
The main brain behind Watson is the POWER 7 processors. IBM has connected several (2,880 of them) together, comprising all of the servers needed to host Watson in order to give him the required mental capacity to process up to 1 million books per second.[1]Software
IBM developed their own software for Watson to run they've coined as, "IBM's DeepQA Software." This system was written in various operating systems to provide the required coding.Data Base
As we will get in to later in this article, Watson's knowledge base is unique to only what he is given. Depending on the course of work being done by Watson, he might be programmed with medical information as an example. He could have dictionaries, scholarly articles from the medical fields, diagnostics information, and just about any other pertinent information that could aid a nurse in determining the best treatment for their patient.Watson's Thought Process
The fundamentals of Watson's operation starts with a question. Once asked, Watson will analyze the question, breaking it down, removing unimportant information. Watson then generates a hypothesis and begins to filter down his data base looking for possible outcomes to the question. Once done, Watson will then review his answers for supporting evidence and start assigning scores to his list of answers. As the list gets weeded out, Watson finally comes to his conclusion and gives a confidence score along with it as to assess how certain Watson is of his work. The following image breaks it down:
The Goal Behind Watson
The ultimate goal behind the creation of Watson was to create a technological system that could process data and give supported answers based on its data base. IBM decided the first test of Watson would be to see if they could create a system that could win a game against the most successful contestants of Jeopardy!
“The challenge is to build a system that, unlike systems before it, can rival the human mind’s ability to determine precise answers to natural language questions and to compute accurate confidences in the answers. This confidence processing ability is key. It greatly distinguishes the IBM approach from conventional search and is critical to implementing useful business applications of Question Answering.” Dr. David Ferucci, Principal Investigator DeepQA Watson Project.[2]
Watson successfully achieved this goal in 2011, so the goal line then moved to address the potential to evolve for commercial use.
References