Scribd
Upload
44 views4 pages

Watson Case Study

Watson is an artificial intelligence system capable of answering questions posed in natural language by analyzing large amounts of stored data. It requires very powerful hardware because it must process millions of documents quickly to classify answers. Although it cannot replace humans in tasks that require intuition, it could be useful as a consulting tool in industries where analyzing large volumes of information is necessary.

Uploaded by

ScribdTranslations
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
44 views4 pages

Watson Case Study

Watson is an artificial intelligence system capable of answering questions posed in natural language by analyzing large amounts of stored data. It requires very powerful hardware because it must process millions of documents quickly to classify answers. Although it cannot replace humans in tasks that require intuition, it could be useful as a consulting tool in industries where analyzing large volumes of information is necessary.

Uploaded by

ScribdTranslations
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

CONCLUSIONS

CASE STUDY: What's up with IBM's Watson?

1. How powerful is Watson? Describe its technology. Why


What does such powerful hardware require?
Watson is so powerful that it can solve problems based on
historical solutions that are stored in its system. Each solution provides
help to define a behavior pattern to respond to such certain types
of questions and additionally the computer Watson offers among its variety of
responses a percentage of reliability for each one.

Regarding the description of the technology, the computer hardware


Watson (used in Jeopardy) consists of 10 IBM server cabinets.
POWER 750 running on Linux. It also has 15 terabytes of RAM and
2,880 processor cores (which is equivalent to 6,000 home computers)
high-end) and operated at 80 teraflops.

Now then, to answer why Watson requires such hardware


powerful, it is due to the fact that the objective of the project was to develop a larger set
effective techniques that computers could use to process the
natural language: the language that human beings use by instinct, not the
language with a special format for computers to understand. To do this
Watson had to be able to record the intention of a question, search
among millions of lines of text and data, choose nuances of meaning and
context, and classify the potential responses that a user could
select, everything in less than three seconds. It should be emphasized that Watson
It contained 10 million documents stored in its main memory
coming from many sources, including encyclopedias, dictionaries,
news articles, newspapers, and literary works.

In conclusion, Watson needs very powerful hardware because it is being


dealing with complex and precise data and information, such as for example
scientific information, and requires the ability to analyze millions of data,
interpret them and offer useful and logical conclusions.
2. How 'intelligent' is Watson? What can it do? What is it?
What can't you do?

Regarding Watson's capabilities, we summarize this in one sentence:


Watson can learn from its mistakes, just as it can learn from its successes. To solve a
common problem, Watson tests many of the thousands of algorithms that he
IBM's team programmed it for use. The algorithms evaluate the language.
used on each track, they collect information about people and places
important ones mentioned in the track and generate hundreds of solutions.

At the same time, if a certain algorithm works to solve a problem,


Watson remembers the type of question it was and the algorithm it used to obtain
the correct answer. In this way, over time Watson improves to
answer questions. Watson also learns in another way: Watson analyzed
both the questions and the answers to determine patterns or similarities
among the clues and, through the use of these patterns, assigns various degrees of
trust in the responses it provides.

Hence, IBM used the term cognitive computing to refer to the


Watson's ability to interpret voice and text, to quickly explore large
data volumes, answer questions, draw conclusions and learn from them
errors.

Finally, the question What can Watson not do? is summarized


in the impossibility of a computer developing significant skills
humanities. Watson has been used in different fields, such as the
search for appropriate treatments for patients with leukemia. In these
cases, Watson could assist specialists, but cannot replace the
skill of an experienced doctor (according to Lynda Chin, president
from the genomics department of M.D. Anderson). Cancer experts have
I have seen patients thousands of times and in some of those instances their decisions are based on
intuition, something that is hard to explain and that no intelligence system
Artificial possesses.

3. What kind of problems can Watson solve? How useful is it?


it is a tool for knowledge management
and decision making?

The types of problems Watson can solve only encompass fields


where small pieces of a knowledge domain can be captured
limited to be used for a single purpose, for example, to respond
correctly answer questions of all kinds from a television program upon having
millions of data and virtual encyclopedias stored. In other words,
Watson helps users solve problems that require
use of large amounts of information in a very short time to
based on prior knowledge and experiences stored in its system.

Watson can become a useful tool in certain aspects because


allows any company to validate and verify information that has already been published.
the moment is unknown, therefore, it can provide information in the least
possible time, which allows saving time and reducing inconsistencies in
the given answers.
On the other hand, in the real world, problems vary in complexity and approach.
since they are not summarized in such a superficial system to face the problems
humans. Our human dimensions range from the bodily dimensions,
affective and cognitive to the aesthetic, moral, and spiritual dimensions between
others.

Finally, despite the aforementioned qualities, Watson presents


it would be difficult to define it as a very useful tool for the
knowledge management and decision-making since it has to
learn the terminology and become an expert in many different areas,
example, medical services and scientific research. It should also
understand the context of how that language is used and how to correlate the
questions with the correct answers. On the other hand, Watson does not work
still with audio, video, and animations data or with other languages apart from
English, this computer still cannot develop its own ideas.

4. Do you think Watson will be as useful in other industries and disciplines?


How does IBM expect? Will it be beneficial for everyone? Explain your
response.

Although Watson's basic learning process has been very arduous and has
required for IBM engineers to master the company's technicalities
a client and translate those requirements into usable software, use this
computers will be beneficial for certain areas of knowledge, as,
in these times of Big Data, the production of content and information is so
voluminous that it becomes impossible to visualize everything, and much less so,
extract from there what is relevant to us; and that is precisely what
make this cognitive system analyze such unstructured information and be able to
provide more accurate results about that information.

Among the examples in favor of the usefulness of Watson, we find the company
Welltok used Watson technology (available from IBM through
Internet as a cloud service to create a mobile application of
CareWell Concierge call for Intelligent Health Itineraries for the
consumers. Now users will be able to engage with Watson in
conversations about their health. The company Fluid Retail is also
developing a personalized shopping assistant and the company MD Buyline
is working on a Watson application to advise managers on
hospitals on the acquisition of medical equipment and supplies.

Now, the text mentions certain favorable movements in these.


aspects as IBM will release three new cloud-based products that
they depend on Watson's cognitive intelligence and capabilities.
IBM Watson Discovery Advisor is geared towards industries
pharmaceutical, editorial and educational, and will consult among the results of
the searches to provide data more quickly and help the
researchers to draw conclusions.
IBM Watson Analytics is a cloud-based service that provides
perspectives, including visual representations, based on the Big
Raw data that companies send to Watson.
IBM Watson Explorer is a cloud service that will provide a view
unified user information to facilitate the disclosure and the
participation of perspectives based on data.

In this way, if we believe that Watson would be useful in other disciplines as long as
when the relevant information regarding the industry is provided
which is desired to be implemented.

Finally, the difficulty of acquiring this technology has become less significant.
with the progressive advances of IBM since the latest version of Watson is
24 times faster than the version used in the Jeopardy contest in
2011, and uses only 10% of the hardware that was used in that version. The costs
they can be more accessible.

You might also like