Upmc tpdev7

© 2016 IBM Corporation
TPDEV
UPMC STL M2 – 2016/2017
Jean-Yves B. Rigolet
IBM Cloud, France Lab
rigolet.j@fr.ibm.com
Développer des applications plus intelligentes
avec Watson

Développer des applications plus intelligentes avec Watson
TPDEV
2
Agenda
The Third Era of Computing
When Computers understood Humans
Discover with Watson
Decide with Watson Analytics
Build with Watson Developer Cloud

TPDEV
Result of IBM Research “Grand Challenge”
On February 14, 2011, IBM Watson made history

TPDEV
1900 1950 2011
. . .enabling new opportunities and outcomes
Watson is ushering in a new era of computing . . .

TPDEV
Watson is a growing tech start-up
R&D
Demonstratio
n
Validation
Research
Project
2006 – 2010
Jeopardy!
Grand
Challenge
2011
Internal
Startup
Division
2011 – 2013
IBM
Watson
Group
2014 – present
Commercializati
on
$1B Investment
$100M for Ecosystem
NYC Headquarters

TPDEV
Mobile Social
Cloud
Internet of Things
Watson
Data is rapidly becoming the foundation for a Smarter Planet

TPDEV
1 in 2
1 in 2
1 in 2
1 in 2
business leaders
don’t have access
to data they need
31.7%
31.7%
31.7%
31.7%
expected CAGR in
the big data
marketplace*
2.2X
2.2X
2.2X
2.2X
more likely that top
performers use
business analytics
80%
of the world’s data
today is
unstructured
90%
of the world’s data
was created in the
last two years
20%
amount of available data
traditional systems
typically leverage
Businesses are “dying of thirst in an ocean of data”

TPDEV
8
Understands
natural language and
human style
communication
Adapts and learns from
training, interaction,
and outcomes
Generates and
evaluates evidence-
based hypothesis
1 2
3
Understands me
Engages me
Learns and improves over time
Helps me discover
Establishes trust
Has endless capacity for insight
Operates in a timely fashion
Watson:
Watson combines transformational capabilities to deliver
a new world experience

TPDEV
Cognitive Systems (‘
Cognitive Systems (‘kognɪtɪv sistem’) - Applying human-like
characteristics to conveying and manipulating ideas, that when
combined with the inherent strengths of digital computing can
help users address problems with higher accuracy, more resilience,
and on a massive scale over very large bodes of information
Watson is a Cognitive System
‒ Teases apart the human language
to identify inferences between text
passages
‒ Can demonstrate human-like accuracy
‒ Performs at speeds far faster, and at
a scale far bigger, than a human
Principles of Cognitive Computing

TPDEV
10
Ushering in the third era of computing

TPDEV
Cognitive computing – Perception
Gaining a new level of insight into our world

TPDEV
Cognitive computing – Reasoning
Drawing inferences to reach new
conclusions

TPDEV
Cognitive computing – Relating
Adapting and personalizing interactions to
each individual

TPDEV
Cognitive computing – Learning
Continuously improving insight with
experience
1K 1M 1B 1T
Interactions
Intelligence

TPDEV
15
Agenda

TPDEV
Informed Decision Making: Search vs. Expert Q&A
Decision Maker
Search Engine
Finds Documents containing Keywords
Finds Documents containing Keywords
Delivers Documents based on Popularity
Delivers Documents based on Popularity
Has Question
Has Question
Distills to 2-3 Keywords
Distills to 2-3 Keywords
Reads Documents, Finds
Answers
Reads Documents, Finds
Answers
Finds & Analyzes Evidence
Finds & Analyzes Evidence
Expert
Understands Question
Produces Possible Answers & Evidence
Delivers Response, Evidence & Confidence
Analyzes Evidence, Computes Confidence
Asks NL Question
Asks NL Question
Considers Answer & Evidence
Considers Answer & Evidence
Decision Maker

TPDEV
Want to Play Chess or Just Chat?
Chess
–A finite, mathematically well-defined search space
–Limited number of moves and states
–All the symbols are completely grounded in the mathematical rules of the game
Human Language
–Words by themselves have no meaning
–Only grounded in human cognition
–Words navigate, align and communicate an infinite space of intended meaning
–Computers can not ground words to human experiences to derive meaning

TPDEV
Easy Questions?
18
ln((12,546,798 * π)) ^ 2 / 34,567.46 =
Owner Serial Number
David Jones 45322190-AK
Serial Number Type Invoice #
45322190-AK LapTop INV10895
Invoice # Vendor Payment
INV10895 MyBuy $104.56
David Jones
David Jones =
0.00885
Select Payment where Owner=“David Jones” and Type(Product)=“Laptop”,
Dave Jones
David Jones
≠

TPDEV
Hard Questions?
Computer programs are natively explicit, fast and exacting in their calculation
over numbers and symbols….But Natural Language is implicit, highly
contextual, ambiguous and often imprecise.
Where was X born?
One day, from among his city views of Ulm, Otto chose a water color to
send to Albert Einstein as a remembrance of Einstein´s birthplace.
X ran this?
If leadership is an art then surely Jack Welch has proved himself a
master painter during his tenure at GE.
Structured
Unstructured

TPDEV
Jeopardy! Basic Game Play
Technology Classics The Great
Outdoors
Speak of the
Dickens
Mind Your
Manners
Before
and After
$200 $200 $200 $200 $200 $200
$400 $400 $400 $400 $400 $400
$600 $600 $600 $600 $600 $600
$800 $800 $800 $800 $800 $800
$1000 $1000 $1000 $1000 $1000 $1000
6 Categories
6 Categories
5 Levels of
Difficulty
5 Levels of
Difficulty
1 of 3 Players Selects a Clue
Host reads Clue out loud
ALL POLICEMEN CAN
THANK STEPHANIE KWOLEK
FOR HER INVENTION OF
THIS POLYMER FIBER, 5
TIMES TOUGHER THAN
STEEL
TECHNOLOGY
All Players compete to answer
1st to buzz-in gets to answer
IF correct
earns $ value
selects Next Clue
IF wrong
loses $ value
other players buzz again
(rebounds)
Two Rounds Per Game + Final Question
ONE Daily Double in First Round, TWO in 2nd Round

TPDEV
21
Broad Domain
Our Focus is on reusable NLP technology for analyzing volumes of as-is text.
Structured sources (DBs and KBs) are used to help interpret the text.
We do NOT attempt to anticipate all questions and build specialized databases.
In a random sample of 20,000 questions we found
2,500 distinct types*. The most frequent occurring <3% of the time.
The distribution has a very long tail.
And for each these types 1000’s of different things may be asked.
In a random sample of 20,000 questions we found
2,500 distinct types*. The most frequent occurring <3% of the time.
The distribution has a very long tail.
And for each these types 1000’s of different things may be asked.
*13% are non-distinct (e.g., it, this, these or NA)
*13% are non-distinct (e.g., it, this, these or NA)
Even going for the head of the tail will
barely make a dent
Even going for the head of the tail will
barely make a dent

TPDEV
Automatic Learning From “Reading”
Officials Submit Resignations (.7)
People earn degrees at schools (0.9)
Inventors patent inventions (.8)
Volumes of Text Syntactic Frames Semantic Frames
Vessels Sink (0.7)
People sink 8-balls (0.5) (in pool/0.8)
Fluid is a liquid (.6)
Liquid is a fluid (.5)

TPDEV
Evaluating Possibilities and Their Evidence
Is(“Cytoplasm”, “liquid”) = 0.2
Is(“organelle”, “liquid”) = 0.1
In cell division, mitosis splits the nucleus & cytokinesis
splits this liquid cushioning the nucleus.
In cell division, mitosis splits the nucleus & cytokinesis
splits this liquid cushioning the nucleus.
Is(“vacuole”, “liquid”) = 0.2
Is(“plasma”, “liquid”) = 0.7
“Cytoplasm is a fluid surrounding the nucleus…”
Wordnet Is_a(Fluid, Liquid) ?
Learned Is_a(Fluid, Liquid) yes.
↑
Organelle
Vacuole
Cytoplasm
Plasma
Mitochondria
Blood …
Many candidate answers (CAs) are generated from many different searches
Each possibility is evaluated according to different dimensions of evidence.
Just One piece of evidence is if the CA is of the right type. In this case a “liquid”.
Many candidate answers (CAs) are generated from many different searches
Each possibility is evaluated according to different dimensions of evidence.
Just One piece of evidence is if the CA is of the right type. In this case a “liquid”.

TPDEV
Keyword Evidence
IBM Confidential
celebrated
India
In May
1898
400th
anniversary
arrival in
Portugal
India
In May
Gary
explorer
celebrated
anniversary
in Portugal
Keyword Matching
Keyword Matching
Keyword Matching
Keyword Matching
Keyword Matching
Keyword Matching
Keyword Matching
Keyword Matching
Keyword Matching
Keyword Matching
24
arrived in
In May, Gary arrived in
India after he celebrated his
anniversary in Portugal.
In May 1898 Portugal celebrated
the 400th anniversary of this
explorer’s arrival in India.
This evidence
suggests “Gary” is
the answer BUT the
system must learn
that keyword
matching may be
weak relative to other
types of evidence

TPDEV
celebrated
May 1898 400th anniversary
arrival
in
In May 1898 Portugal celebrated
the 400th anniversary of this
explorer’s arrival in India.
Portugal
landed in
27th May 1498
Vasco da Gama
Temporal
Reasoning
Statistical
Paraphrasing
GeoSpatial
Reasoning
explorer
On the 27th of May 1498, Vasco da
Gama landed in Kappad Beach
Kappad Beach
Para-
phrase
s
Geo-
KB
Date
Math
25
India
Stronger
evidence can
be much
harder to find
and score. The evidence is still not 100% certain.
Search Far and Wide
Explore many hypotheses
Find Judge Evidence
Many inference algorithms
Deeper Evidence

TPDEV
Not Just for Fun
A long, tiresome speech delivered by a frothy pie topping
Answer:
Meringue
Harangue
Harangue
Meringue
.
Diatribe
.
.
.
.
Whipped Cream
.
.
.
.
.
Category: Edible Rhyme Time
26
Some Questions require
Decomposition and Synthesis

TPDEV
In 1968
When "60 Minutes" premiered this man was U.S. president.
this man was U.S. president.
Lyndon B Johnson
?
27
Divide and Conquer
(Typical in Final Jeopardy!) Must identify and solve
sub-questions from different
sources to answer
the top level question

TPDEV
The Missing Link
On hearing of the discovery of George Mallory's body, he told reporters he still thinks he was first.
TV remote controls,
Buttons
Shirts, Telephones
Mt
Everest
He was first
Edmund
Hillary

TPDEV
29
The Best Human Performance: Our Analysis Reveals the Winner’s
Cloud
Winning Human
Performance
Winning Human
Performance
2007 QA Computer System
2007 QA Computer System
Grand Champion
Human Performance
Grand Champion
Human Performance
Each dot represents an actual historical human Jeopardy! game
More Confident Less Confident

TPDEV
The Technology Behind Watson
Massively Parallel Probabilistic Evidence-Based Architecture
Generates and scores many hypotheses using a combination of 1000’s Natural Language Processing,
Information Retrieval, Machine Learning and Reasoning Algorithms.
These gather, evaluate, weigh and balance different types of evidence to deliver the answer with the best support
it can find.
. . .
Answer
Scoring
Models
Answer &
Confidence
Question
Evidence
Sources
Models
Models
Models
Models
Models
Primary
Search
Candidate
Answer
Generation
Hypothesi
s
Generatio
n
Hypothesis and
Evidence Scoring
Final
Confidence
Merging &
Ranking
Synthesi
s
Answer
Sources
Question
& Topic
Analysis
Evidence
Retrieval
Deep
Evidence
Scoring
Learned Models
help combine and
weigh the
Evidence
Hypothesis
Generation
Hypothesis and Evidence
Scoring
Question
Decomposition
1000’s of
Pieces of Evidence
Multiple
Interpretations
100,000’s Scores from
many Deep Analysis
Algorithms
100’s
sources
100’s Possible
Answers
Balance
& Combine

TPDEV
Grouping Features to produce Evidence Profiles
Clue: Chile shares its longest land border with this country.
Clue: Chile shares its longest land border with this country.
Bolivia is more
Popular due to a
commonly
discussed
border dispute..
Positive Evidence
Positive Evidence
Negative Evidence
Negative Evidence

TPDEV
Evidence: Time, Popularity, Source, Classification, etc.
Clue: You’ll find Bethel College and a Seminary in this “holy” Minnesota city.
There’s a Bethel College and a Seminary in both
cities. System is not weighing location evidence
high enough to give St. Paul the edge.

TPDEV
Evidence: Puns
Humans may get this based on the pun since St.
Paul since is a “holy” city. We added a Pun Scorer
that discovers and scores Pun relationships.

TPDEV
St. Petersburg is home to
Florida's annual
tournament in this game
popular on shipdecks
(Shuffleboard)
St. Petersburg is home to
Florida's annual
tournament in this game
popular on shipdecks
(Shuffleboard)
U.S. CITIES
U.S. CITIES
Rochester, New York
grew because of its
location on this
(the Erie Canal)
Rochester, New York
grew because of its
location on this
(the Erie Canal)
From India, the shashpar
was a multi-bladed
version of this spiked club
(a mace)
From India, the shashpar
was a multi-bladed
version of this spiked club
(a mace)
Country Clubs
Country Clubs
A French riot policeman
may wield this, simply the
French word for "stick“
(a baton)
A French riot policeman
may wield this, simply the
French word for "stick“
(a baton)
Archibald MacLeish?
based his verse play "J.B."
on this book of the Bible
(Job)
Archibald MacLeish?
based his verse play "J.B."
on this book of the Bible
(Job)
Authors
Authors
In 1928 Elie Wiesel was
born in Sighet, a
Transylvanian village in
this country
(Romania)
In 1928 Elie Wiesel was
born in Sighet, a
Transylvanian village in
this country
(Romania)
Watson uses statistical machine learning to discover that Jeopardy!
categories are only weak indicators of the answer type.
Categories are not as simple as the seem

TPDEV
In-Category Learning CELEBRATIONS
OF THE MONTH
CELEBRATIONS
OF THE MONTH
What the Jeopardy! Clue is asking for is NOT always obvious
Watson can try to infer the type of thing being asked for from the previous answers.
In this example after seeing 2 correct answers Watson starts to dynamically learn a
confidence that the question is asking for something that it can classify as a “month”.
Clue Type Watson’s Answer Correct Answer
D-DAY ANNIVERSARY & MAGNA
CARTA DAY
day Runnymede June
NATIONAL PHILANTHROPY DAY &
ALL SOULS' DAY
day Day of the
Dead
November
NATIONAL TEACHER DAY &
KENTUCKY DERBY DAY
Day/month(.2) Churchill
Downs
May
ADMINISTRATIVE PROFESSIONALS
DAY & NATIONAL CPAS GOOF-OFF
DAY
day / month(.6) April April
NATIONAL MAGIC DAY & NEVADA
ADMISSION DAY
day / month(.8) October October

TPDEV
Baseline 12/06
v0.1 12/07
v0.3 08/08
v0.5 05/09
v0.6 10/09
v0.8 11/10
v0.4 12/08
v0.2 05/08
IBM Watson
Playing in the Winners Cloud
V0.7 04/10
Incremental Progress in Answering Precision on the
Jeopardy Challenge: 6/2007-11/2010

TPDEV
Betting
Game Strategy
Managing the Luck of the Draw
IBM Confidential
Dynamic
Confidence
Clue
Selection
Risk
Management
Modulate
Aggressiveness
Direct Impact
on Earnings
DDs & FJ
Learn at lower
risk
Prior
Probabilities
Performance
Relative to
Competition
Common
Betting
Patterns

TPDEV
One Jeopardy! question can take 2 hours on a single 2.6 Ghz
Core
Question
100s Possible
Answers
1000’s of
Pieces of Evidence
Multiple
Interpretations
100,000’s scores from many simultaneous
Text Analysis Algorithms
100s sources
. . .
Hypothesis
Generation
Hypothesis and
Evidence Scoring
Final Confidence
Merging &
Ranking
Synthesis
Question &
Topic
Analysis
Question
Decomposition
Hypothesis
Generation
Hypothesis and Evidence
Scoring
Answer &
Confidence
Watson had to answer in 2-6 seconds.

TPDEV
Deep Analytics – Combining many analytics in a novel
architecture, we achieved very high levels of Precision and
Confidence over a huge variety of as-is content.
Speed – By optimizing Watson’s computation for
Jeopardy! on over 2,800 POWER7 processing cores we
went from 2 hours per question on a single CPU to an
average of just 3 seconds.
Results – in 55 real-time sparring games against former
Tournament of Champion Players last year, Watson put
on a very competitive performance in all games -- placing
1st in 71% of the them!
Precision, Confidence & Speed

TPDEV
40
Cross-industry
Watson Engagement Advisor
Watson Discovery Advisor
Watson Analytics (NEO)
Watson Explorer
Industry-specific
Watson Health Cloud
Watson Travel Cloud
Watson Retail Cloud
Etc….
Ecosystem
Watson Developer Cloud
Watson Content Store
Watson Talent Hub
Watson Products and
Services
Business solutions
and services
• IBM Decision Management
• IBM Content Analytics
• IBM Planning & Forecasting
• IBM Discovery & Exploration
• IBM BI & Predictive Analytics
• IBM Content Management
• IBM Hadoop System
• IBM Stream Computing
• IBM Data Management
& Data Warehouse
• IBM Information Integration
& Governance
Watson Foundations
New era big data and analytics
capabilities
The progression forward in 2013:
Feb – Wellpoint delivered two
healthcare solutions
powered by IBM Watson
May – announced Watson
Engagement Advisor to
transform how brands
and their customers
interact over the lifetime
of their relationship
October – MD Anderson and
Cleveland Clinic both
announced their work
with Watson to help with
oncology and medical
education (respectively)
Watson-enabled Solutions
(IBM & Client)
IBM and client solutions that
incorporate Watson services
Systems & storage optimized for big
data & analytics
Systems & Storage
Introducing the IBM Watson family

TPDEV
41
Agenda

TPDEV
“Medicine has become
too complex. Only about
20% of the knowledge
clinicians use today is
evidence-based.”
Steven Shapiro
Steven Shapiro
Steven Shapiro
Steven Shapiro
Chief Medical & Scientific Officer
University Pittsburgh Medical
Center
Healthcare presents some formidable challenges

TPDEV
Analytics innovations
Advanced analytics like IBM
Watson apply reasoning, and
complement tools that surface
patterns and anomalies
New
platforms
Collaborative and
mobile technology
platforms facilitate a
holistic view of the
individual and
enable new ways to
coordinate care
delivery
Rich data
sources
EHRs, Labs, Genetic
Biomarkers, Bio-
Monitoring, New Targeted
Therapies, Claim History,
Family Histories, Patient
Preferences
Evidence-Based, Precision,
Personalized, Patient-
Centered Medicine
Rich data and advanced technologies are fueling change
in healthcare

TPDEV
44
Value:
Increased R&D
productivity
Faster time to insight
Improved visibility to
external and internal
research
Increased currency /
relevancy of analysis
User: Researchers, Analysts
Watson can read these medical records
in six seconds!
Watson Discovery Advisor enables researchers to
answer the tough research problems that have never been
answered before

TPDEV
45
Agenda

TPDEV
Making decisions rapidly
is no longer a goal; it’s an
imperative
The desire to make data-
driven decisions is
prevalent
Access to required data
sources is critical while
maintaining governed
standards
34% can not find
time to analyze
data
38% have a
limited
understanding of
how to use
analytics
24% find it
difficult to get
data
Leveraging analytics still faces many obstacles

TPDEV
Data Access
Data Preparation
Analysis
Validation
Collaboration
Reporting
Business
Analysts
Business
Users
Data Scientists
and
Statisticians
IT
Even a simple analytics project has multiple steps and
people

TPDEV
And it’s rarely a straightforward process
Business
Users
Data Scientists
and
Statisticians
Business
Users
Data Scientists
and
Statisticians
IT
Data Access
Analysis
Validation
Collaboration
Reporting
Data Preparation
Business
Analysts
And it’s rarely a straightforward process

TPDEV
Put analytics in the hands of everyone
Make data access and refinement easier
Deliver through the cloud for agility and speed
Think Ahead Tell a Story
Understand Your
Business
Get Better Data
Mobile Ready Secure
Integrated information services
provide data access and
refinement
Automated intelligence
accelerates your ability to
answer questions
Guided predictive analytics
reveals insights and
opportunities
Visualizations support your
decisions and communicate
results
If only you could...

TPDEV
Business Users Business Analysts Data Scientists IT
Self-service analytics for business users and experts alike
Prioritizing
Accounts
Receivable
Employee
Retention
Helpdesk
Case
Analysis
Campaign
Planning and ROI
Warranty
Analysis
Customer
Retention

TPDEV
Natural
language
dialogue
Data
discovery
Quick start
intuitive
interface
Mobile-ready
Cloud-based agility

TPDEV
Visual
storytelling
Intelligent
automation
Data access
and
refinement
Report and
dashboard
creation
Integrated
social
business
Guided
analytic
discovery
Unified analytics experience

TPDEV
53
Agenda

TPDEV
Access restricted to partners
and IBM developers
Accessible by anyone with a
Bluemix account
Wait until services are GA to
release
Release in Beta and gather
input from user community
North America Global
One service Eight services and more coming
© 2014 International Business Machines Corporation
IBM has radically expand access to Watson services

TPDEV
55
20 Watson Services for Bluemix available today
with more to come, stay tuned!
Each service performs a unique
task
Cognitive categories
Language
Speech
Vision
Data Insights

TPDEV
All Watson Services offer REST APIs

TPDEV
… And how to use the service (including sample apps)

TPDEV
Understands question
Produces possible answers and
evidence
Analyzes evidence
Computes confidence
Delivers response, evidence and
confidence
Asks a question
Considers response and
evidence
Ask a question and get an answer with
supporting evidence
How does it work?
Interprets and answers user questions directly based on
primary data sources (brochures, web pages, manuals,
records, etc.) that have been selected and gathered into a
body of data or ‘corpus’. The service returns candidate
responses with associated confidence levels and links to
supporting evidence.
Use Cases
Healthcare: What is a stroke? What is the cause of
Wilson Disease?
Travel: Where is the best place to stay in Prague?
Input: Your questions.
Output: List of passages and confidence scores
Question and Answer

TPDEV
Uncover a deeper understanding of people's personality
characteristics, needs, and values to drive personalization
How does it work?
Personality Insights extracts and analyzes a spectrum of personality
attributes to help discover actionable insights about people and entities,
and in turn guides end users to highly personalized interactions. The
service outputs personality characteristics that are divided into three
dimensions: the Big 5, Values, and Needs. While some services are
contextually specific depending on the domain model and content,
Personality Insights only requires a minimum of 3500+ words of any
text.
Use Cases
The service can analyze text based on a customer’s twitter stream to
help a travel agency decide between leading with a budget or luxury
trip offer
Anywhere improving a customer engagement can help create an
organization differentiate itself.
Input:
Output: Portrait API returns a JSON
visualization.
Visualization API a visual HTML and
SVG
Personality Insights

TPDEV
Translate and publish content in multiple
languages
How does it work?
The Watson Language Translation service provides
domain-specific translation utilizing Statistical Machine
Translation techniques that have been perfected in our
research labs over the past few decades. Currently, three
domains are available that provide translation among a
total of seven languages. For best results, a domain that
matches the content to be translated should be chosen.
Use Cases
A French speaking help desk representative is assisting a
Portuguese speaking customer through a chat session
and is able to interact through the translation service
Input: plain text
Output: plain text in selected language
Language Translation

TPDEV
Rules-based user response system. Script conversations
any way you like to answer questions, walk through
processes, or just to chat!
How does it work?
The IBM Watson Dialog service enables a developer to automate
branching conversations between a user and your application. The
Dialog service enables your applications to use natural language to
automatically respond to user questions, cross-sell and up-sell,
walk users through processes or applications, or even hand-hold
users through difficult tasks. The Dialog service can track and store
user profile information to learn more about end users, guide
them through processes based on their unique situation, or pass
their information to a back-end system to help them take action
and get the help they need.
Use Cases
Walk a user through the steps to reset their password or help
them choose a credit card. Developers script conversations as they
would happen in the real world, upload them to the Dialog
application and allow end users to get the help they need
Dialog

TPDEV
Link euphemisms or colloquial terms to more
commonly understood phrases
How does it work?
Concept Expansion is a Watson service that analyzes large
amounts of text to create a dictionary of contextually
related words. Concept Expansion's pattern recognition
technology helps users identify contextually similar terms
and phrases, create dictionaries, and find or organize text
based on those dictionaries. It knows that 'The Big Apple'
refers to New York City and that 'getting in touch' means
communicating by email, letter, or phone.
Use Cases
Concept Expansion can be used to build dictionaries of
conceptually related words across large amounts of
unstructured data in multiple domains such as financial
services, legal, medical etc.
Input: List of terms to
start seed list and a fixed
fata set.
Output: Ranked list of similar terms found
within the fixed data set
Concept Expansion

TPDEV
Extracts relationships between different text
entities
How does it work?
Unlike general-purpose text analytics tools, Relationship
Extraction leverages Watson machine learning
technologies. The API can analyze news articles and use
statistical modeling to perform linguistic analysis of the
input text. It then finds spans of text and clusters them
together to form entities, before finally extracting the
relationships between them.
Use Cases
Relationship Extraction can be used to build
applications such as semantic search engines, business
and competitive intelligence tools, knowledge graphs,
social media monitoring tools, and more.
Can be used in business and competitive intelligence
to find information about competitors’ products in the
news
Useful for brand management; helps find information
about your own products
Input: News articles.
Output: XML document displaying the relationships
Relationship Extraction

TPDEV
Watson Content Marketplace
http://www.ibm.com/smarterplanet/us/en/ibmwatson/developercloud/content-marketplace.html

TPDEV
65

Upmc tpdev7

Recommended

Recommended

More Related Content

Similar to Upmc tpdev7

Similar to Upmc tpdev7 (20)

More from Jean-Yves Rigolet

More from Jean-Yves Rigolet (10)

Recently uploaded

Recently uploaded (20)

Upmc tpdev7