Evaluation of International Business Machines Corporation's Watson

Summary

Case Study in Week 5 talks about Watson, an intelligence system created by IBM. In early spring of 2011, the International Business Machines Corporation (IBM) had put their computer intelligence system, Watson to the test. The test would prove Watson’s ability to interpret data as well as recognizing patterns in human language and to make decisions on how to answer questions accordingly. Watson took on Jeopardy champions Ken Jennings and Brad Rutter. The concept was developed as a challenge for a computer to recognize and interpret natural spoken language rather than deliberately formatted language used in thought-out speech or writing (Laudon & Laudon, 2016).
 Watson’s development and speed allowed it to process language, consider key elements picked up through the speech and then give up to hundreds of acceptable responses. An intriguing element of Watson’s programming was its similarity to the human brain in that over time, Watson’s intelligence for answering questions would improve. Perhaps, Watson’s method of improvement was different since the computer would record and recognize algorithms related to the scheme of the types of questions it has been asked (Laudon & Laudon, 2016).
 Initially, Watson was intended to help improve applications for healthcare, financial services, education and other areas which obtain large amounts of data. Six months following Watson’s Jeopardy challenge, the computer system was added to WellPoint Inc., the largest U.S healthcare provider as a component for management in assisting physicians and their patients in selecting the most optimal cancer treatment per individual. Watson underwent programming that would provide the computer with respective profiles of patients, paired with the understanding of medical features such as treatment type and dosage; then the computer would select a treatment method most ideal for the present patient. Of course, a computer could never replace a working physician; however, today Watson is accessible for business companies as a cloud-based product with capabilities ranging from pharmaceutical departments to education and publishing industries. There is no saying just how far this idea can go, however it is likely that we are still far-off from the conceptualized world ran by computers (Laudon & Laudon, 2016).
 Team 8 will support our findings to questions such as whether Watson is powerful or not, how intelligent Watson may be, what problems Watson may be able to resolve, and if we determine whether or not Watson may be useful in other industries and disciplines.
Support
International Business Machines Corporation (IBM) operates around the world marketing hardware, software, plus offers consulting services for mainframe computers. IBM holds many patents in research for business operations. They created the ATM, floppy disks, UPC barcodes, SQL programming language, and the first IBM mainframe in the 1960’s-1970 was the dominant computing platform during that time (Madrigal, 2011).
According to an article in Forbes (Press, 2016), in 1952 IBM released its first example of artificial intelligence when Arthur Samuel developed the first computer to play checkers and learn from each game. In 2011, 14 years later, IBM introduced another artificial intelligence program Watson, named after Thomas Watson, the first CEO/President of IBM (Bracetti, 2015).
Watson is a technology platform that uses machine learning and language processing to mine large amounts of data to answer questions, make comparisons, and deliver usable information in variable forms (Press, 2016).
Since IBM made Watson available through cloud services in 2013, many companies have been able to take advantage of its capabilities. In 2014, IBM Watson and Genesys became partners to help organizations and consumers connect through self-service applications to better resolve customer issues or concerns by allowing for responses to be delivered within a few minutes (M2 Presswire, 2014).

Other types of industries have also begun to capitalize on the use of Watson. In healthcare, Watson is being used for planning patient treatment options in oncology. Different treatments are proposed, and then the Medical Team makes the decisions on whether the treatments proposed are applicable or not. In education, Sesame Workshop and IBM are using Watson’s pattern recognition and language processing skills to create preschool curriculums based on children’s learning styles (Fast Company, 2017).
It will be interesting to see if Watson can further adjust and provide educational curriculums for children along the Autism spectrum, as that is an area that is extremely difficult for teachers due to the extreme sensitivity these children have to stimulation. Furthermore, the general manager of IBM Watson David Kenny states that “Our mission is to (eventually) let people own their own AI” (Fast Company, 2017). This would be very significant to assisting young children with learning disabilities. .             
Evaluation
Is Watson effective artificial intelligence? Our book describes three types of data systems to help users. We will discuss two that closely relate. They are expert systems and case-based reasoning. Watson is like many other artificial intelligence systems where it needs be taught, and learn from its mistakes (Laudon & Laudon, 2016, p. 443).
 An expert system is one that is programmed specifically and deliberately with rules. It is programmed with many if-then-else scenarios, and does not deviate from the rules. Once all of the rules have been exhausted, you have your answer. This is not exactly how Watson is built; it has data fed into it, and it searches the data to make a decision. However, it does have the same limitations where it can only decide from the information that has been inputted, and it needs experts to help define what answers are good or not, and be refined over time (Laudon & Laudon, 2016).
 Case-based reasoning is more of what IBM’s Watson is built to do. Watson is fed past situations, problems, what solutions were tried, and the results. It will analyses the parameters of the case students to be able to identify patterns with future input. From the input, it will search for successful outcomes for problem sets in the past, and give recommendations of solutions to try. It is not always right, so it gives a degree of certainty with each answer (Laudon & Laudon, 2016).
 The applications are still limited because it takes a lot of time and money to create the data, and relationships needed to be effective in an industry. Watson doesn’t just search the internet for answers, it has to have information picked and loaded. In this way, the implementations have been limited to those of expert systems (Laudon & Laudon, 2016).
Case Study Questions
11-13 How powerful is Watson? Describe its technology. Why does it require so powerful hardware?
Watson is the computer that was built by IBM Corporation to achieve something that no other computer has done before in its field. Watson has the ability to process and interpret human language. The goal of Watson’s project was to develop a computer, which can process human beings natural language using a more effective set of techniques. The database of Watson consists of a large amount of data. The team downloaded millions of documents from internet such as Wikipedia and encyclopedia information and uploaded in Watson’s primary memory; therefore, Watson requires a significant amount of data to successfully scan huge quantity of data present in its database quickly (Laudon & Laudon, 2016).

11-14 How “intelligent” is Watson? What can it do? What can’t it do?
Watson can currently perform solutions for only fraction of problems but it cannot guess solutions for problems due to running out of memory. Watson is expected to become more useful and gain more knowledge using more sets of questions from different fields. It’s not good enough to predict human intention and feelings. Watson consists of ten racks of IM Power 750 servers running Linux, and uses 15 terabytes of RAM. It also uses 2,880 processor cores which are equivalent to 6,000 modern computers. It also operates at 80 teraflops. The reason why Watson needed all this power is because it had to quickly scan its huge database of information, including the one from the Internet.  (Laudon & Laudon, 2016).
11-15 What kinds of problems is Watson able to solve? How useful a tool is it for knowledge management and decision making?
Since, it’s much like humans, it can learn from its mistakes and successes. Watson consists of thousands of algorithms, to provide solutions for a typical problem. Watson remembers a certain algorithm, which works to solve a type of question, and uses it next time to get the right answer. IBM has invested in its artificial intelligence device to develop applications for financial services, healthcare services or any other industry. It is still unclear how Watson can be effective in generating solutions related to medical fields. The doctors understand about a disease in human body with the help of knowledge, good guesses and past experience; whereas this is something Watson has not fully mastered yet (Laudon & Laudon, 2016).
IBM knows that Watson cannot replace human doctors, but it can help to improve medical diagnostics and fatal medical errors using its advanced analytics technology. Watson has the ability to read and process human beings natural language. It is used as a tool for decision-making and knowledge management processes. It has the ability to perform various jobs in business organizations, which requires real knowledge about that job. It can also become an expertise in performing that job. Watson’s computer models are used to forecast weather conditions, to provide feedback for online shopping sites (Laudon & Laudon, 2016).
11-16 Do you think Watson will be as useful in other industries and disciplines as IBM hopes?
Will it be beneficial to everyone? Explain your answer.
IBM’s Watson is an artificially intelligent computer system capable of answering questions posed in language. It can observe, interpret, evaluate and decide. Use of artificial intelligence can be both useful as well as destructive. It is like different sides of the same coin. According to IBM, Watson is the new species of artificial intelligence and it will be beneficial to multiple industries in multiple ways. Watson may not be beneficial to everyone as it will create problems such as increase unemployment in developing countries, increased support on artificial intelligence can lead to more fraud, etc. Watson can go through the process of trial and error at a very past pace and may be able to keep track of methods and algorithms that cannot provide the right answer; and hopefully identify this in order to get to the correct one in the future (Laudon & Laudon, 2016).

References

• Bracetti, A. (March 6, 2105) 20 Things You Didn’t Know About IBM. Taken from:
  https:www.thrillist.com/tech/ibm-trivia-history-and-things-you-didn-t-know#
• Fast Company (March 2017). IBM for Imbedding Watson Where it is Needed Most. Fast
  Company. March 2017, Issue 213, p54.
• Laudon, K. C., & Laudon, J. P. (2016). Management Information Systems – Managing the Digital Firm (15th ed.). Upper Saddle River: Pearson Education Inc.
• Madrigal, A. (2011, Jun 16). IBM’s First 100 Years: A Heavily Illustrated Timeline. Retrieved from The Atlantic: https://www.theatlantic.com/technology/archive/2011/06/ibms-first-100-years-a-heavily-illustrated-timeline/240502/
• M2 Presswire, Coventry (June 12, 2014). IBM Watson and Genesys to Power Smarter Customer Experiences; Cognitive Learning and Customer Experience Leaders will transform the Way Consumers and Organizations Interact. ProQuest.
• Press, G. (2016, Dec 30). A Very Short History of Artificial Intelligence. Retrieved from Forbes: https://www.forbes.com/sites/gilpress/2016/12/30/a-very-short-history-of-artificial-intelligence-ai/#69bf4e846fba