2. HYBRID
TEXT
ANALYTICS
Combines AI with rule-based
knowledge structures (created
by some very clever humans).
Because AI is still too dumb
to understand the real
meaning in human
language.
4. In a nutshell….
Annual/Pulse
Surveys
Continuous
Listening
WWW
Emotions & Behavioural Trends
Understand why people do what they do
Make better products
Be a better employer
Grow your organisation
Innovate & Improve
Make better decisions
The Hybrid Text Analytics
engine extracts…
…which you
use to..
6. Why the use of Artificial
Intelligence is limited in
Text Analytics.
7. This means NLP can’t extract any real value from text data.
There is a Fundamental Problem
Artificial Intelligence isn’t
smart enough to understand
human language
Natural Language Processing (NLP – a component of AI)
tries to understand and classify language.
But NLP can’t identify word context, meaning,
or relevance.
8. Machine Learning (ML) is another component of AI
But teaching the Machine is a huge task
ML is used to identify themes, positivity, and
sentiment within data sets.
In order to Learn, the Machine needs a vast quantity
of professionally annotated comments to confidently
predict the meaning in an extract of text.
Teaching the Machine is an unviable solution.
9. … modern deep learning-based NLP
models see benefits from much larger
amounts of data, improving when trained
on millions, or billions, of annotated
training examples.”
Google AI blog
Neither provide any deep
value extraction or Business
Intelligence capability.
“AI is poor at dealing with unknown and
unstructured spaces(text).”
Kai-Fu Lee
“Don’t fall for the hype that AI will solve all
of your text analytics needs. Just the
opposite; in this evaluation we found that
rules still rule.
Mostly rules-based text analytics
platforms are much more accurate out of
the box and require much less training
than platforms based mostly on machine
learning.”
Forrester 2019 Text Analytics Platforms
report
These are two AI solutions in
the Text Analytics industry.
10. The secret to releasing value from unstructured
text data is using a Hybrid Text Analytics (HTA) engine
A HTA engine integrates rule-based Keyword and Ontology
sets with components of AI, leveraging human-defined
coding structures with scalable and cutting-edge technology.
Keywords are complex constructions. Hundreds of
conditionalities manually wrapped around each word and
phrase that enable AI to understand language.
Ontologies are manually crafted data coding structures that
ensure meaning and relevance of the language can be assessed
and analysed by the AI.
11. Combining Keyword and Ontology sets with cutting-edge NLP and ML
creates a Hybrid Text Analytics engine that understands the
sentiment, meaning, and emotions in text data.
#HTA #scalable #highaccuracy #highspeed #API #SaaS #insightsonaplate #BusinessGold
Every piece of your unstructured text data
can be effectively mined for game-changing
Business Intelligence.
12. Pansensic has integrated our Hybrid Text Analytics engine within the
ecosystems of some of the world’s largest organisations, and is using it to
provide SaaS capability and consultancy to our global partners and clients.
Use it to gain previously unknown advantage by understanding your
employees, consumers, competitors, and investment markets.
Contact us for an informal walk-through of the capability
+44 (0) 203 432 9804
insight@pansensic.com