Scribd
Upload
119 views22 pages

Ibm Watson

The document discusses using natural language processing and IBM's Watson supercomputer to gain relevant insights from unstructured data like news feeds. Specifically, it aims to focus on efficient reading and capturing insights. It describes how Watson was created using advanced natural language processing, information retrieval, knowledge representation, and machine learning technologies to function as a question answering system. It then provides details on using IBM's Watson Knowledge Studio and its natural language understanding service to build custom models combining AI and machine learning with data analysis.

Uploaded by

vivek2270834
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
119 views22 pages

Ibm Watson

The document discusses using natural language processing and IBM's Watson supercomputer to gain relevant insights from unstructured data like news feeds. Specifically, it aims to focus on efficient reading and capturing insights. It describes how Watson was created using advanced natural language processing, information retrieval, knowledge representation, and machine learning technologies to function as a question answering system. It then provides details on using IBM's Watson Knowledge Studio and its natural language understanding service to build custom models combining AI and machine learning with data analysis.

Uploaded by

vivek2270834
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 22

NITANSHI MAHAJAN

What if we could gain relevant insights from our news


feed so that we only read articles or news about
something/someone we are concerned about ?
Aim of the Project :
1) Focus on efficient reading.
2) Capture insights from unstructured data.

How to deal with this problem ?


Using the Natural Language Processing.
An IBM supercomputer that
combines artificial intelligence
(AI) and sophisticated
analytical software for optimal
performance as a “question
answering” machine.
Watson was created as a question
answering (QA) computing system that
IBM built to apply -

Advanced natural language processing


 Information retrieval
 Knowledge representation
 Automated reasoning
Machine learning technologies
 Cloud – based application.
 Helps in building custom made models.
 Combines AI and ML with data analysis.
Mention detection
Relation Detection
Co reference Resolution
Rule Definition
Regex Definition
Other functionality
PRE - ANNOTATION

TRAINING AND ANALYSIS


TRIAN IBM
MODEL DEPLOY WATSON
ON IBM TO NLU
WKS SERVICE
Uses Natural Language Understanding(NLU) techniques
that help computer understand the complexities of
human language to derive fast and meaningful insights
from large unstructured data.

It helps in drawing various parameters such as :


Concepts
Emotions
Categories
Relations
Entities
Keywords
Semantic Roles
Sentiment
There was a small test model built by me to understand the
working of IBM WKS and it’s NLU service to get a gist of how
ML works.
Created workspaces of domains for which the
models were to be built.
Uploaded documents in the formats specified.
Creating entity types and relation types for
annotation purpose. In order to accelerate the
process, we could use pre- annotators.
After defining entity types , relation types and other
things, we annotate our document.
To help Watson understand nuances of the language of
our domain, we train Watson using defined rules.
After the training is done, we test and evaluate our
model and iterate the annotation process with more
accuracy until we obtain optimum results.
After training and evaluation, we deploy our model
to one of the Watsons’ Services to find solution to our
business problem.
Natural Language Processing (NLP) is the art and
science which helps us extract information from
text and use it in our computations and algorithms.

Named Entity Sentiment


Stemming
Recognition Analysis

Named Entity Semantic Text


Lemmatization
Disambiguation Similarity

Word Language
Part of Speech
Embedding Identification
Autocorrect
Spellchecker
Grammar Checker
Spam Detection
Email classification in Gmail
Duplicate detection
And many more…

You might also like