SlideShare a Scribd company logo
1 of 50
Bangalore, India
A Platform for Text Analytics
Vaklipi
http://www.vaklipi.com
Aiaioo Labs vaklipi.com
Aiaioo Labs
• Research Lab
– Natural Language Processing
– Machine Learning
– Information Retrieval
• Vaklipi
– Vaklipi is the brand under which our text
processing tools are distributed.
– Vak means Speech or Spoken Language
– Lipi means Writing or Written Language
Aiaioo Labs vaklipi.com
What can Vaklipi help you get out of text?
The opportunities vary by vertical:
1. Aerospace / Defense / Automotive – Compliance process automation /
Technical specification standardization / Competitive intelligence and
customer feedback processing
2. Healthcare / Life sciences – Reporting / Storing, retrieving and analyzing
patents and publications / Extraction of actionable insights from research
and competitive intelligence
3. Legal and Government – Legal and administrative filings / Case document
and administrative record management, e-discovery / Extraction of facts
from legal and administrative documents (land records, case files)
4. FMCG – Routine reports, invoicing, tax documents / Sentiment and
Intention Analysis / Call center insights
5. BFSI – Reporting / Churn and product propensity analysis / Forms
processing, customer care, monitoring, fraud detection
Aiaioo Labs vaklipi.com
How do we unlock
the value in “text”?
Output Text / Store and Analyse Text / Ingest Text
d. Vaklipi Generation Tools
a. Vaklipi Extraction Tools
b. Vaklipi Classification Tools
c. Vaklipi Exploration Tools
Aiaioo Labs vaklipi.com
Vaklipi Extraction Tools
“Extraction”
Aiaioo Labs vaklipi.com
a. Vaklipi Extraction Tools
Use Cases:
1. Customer Service
2. Land Records
3. Mergers and Acquisitions
4. Sentiment Analysis
5. Intention Analysis
6. Sales Lead List Extraction
(4, 5 and 6 are supported out of the box, while
1, 2 and 3 will require customization)
Aiaioo Labs vaklipi.com
Use Case 1: Customer Service
Let’s say you have some text … … and a database or spreadsheet with columns
“John Chambers of Springfield, MA
reported a problem with the clutch
on his Ford Ranger purchased in
Boston, MA in 2005.”
… and you have to fill in the database fields
from the information in the text …
Reporter Location (of
Reporter)
Product
Aiaioo Labs vaklipi.com
Use Case 1: Customer Service
“John Chambers of Springfield, MA
reported a problem with the clutch
on his Ford Ranger purchased in
Boston, MA in 2005.”
Entities are pieces of text that could go into the fields in the database.
Relations connect the entities that belong in a row.
Reporter Location Product
John
Chambers
Springfield,
MA
Ford
Ranger
Aiaioo Labs vaklipi.com
Use Case 2: Government – Land Records
Let’s say you have some text … … and a database or spreadsheet with columns
“Property K45L234
(lot 23-24) in Wake County
of 3000 sq ft
was sold to James Fischer
on 3-30-1997 …”
… and you have to fill in the database fields
from the information in the text …
Title# Lot County
Aiaioo Labs vaklipi.com
Use Case 3: Legal Domain – Mergers/Acquisitions
Let’s say you have some text … … and a database or spreadsheet with columns
“Acme Financials, a subsidiary
of Lehman Sisters, was acquired
by John Doe Corp on 5/26/2001.”
… and you have to fill in the database fields
from the information in the text …
Acquirer Acquired Date
Aiaioo Labs vaklipi.com
Use Case 4: Sentiment Analysis
Let’s say you have some text … … and a database or spreadsheet with columns
“The Canon 3F model is sturdy
and excellent for studio use.”
… and you have to fill in the database fields
from the information in the text …
Product Sentiment Purpose
Canon 3F Positive Studio
Use
Aiaioo Labs vaklipi.com
Use Case 4: Sentiment Analysis
Live demo: http://www.aiaioo.com:8080/annotator-0.1/automation/demoView/1
Aiaioo Labs vaklipi.com
Use Case 4:
Sentiment Analysis is available in many languages
Live demo: http://www.aiaioo.com:8080/annotator-0.1/automation/demoView/1
Aiaioo Labs vaklipi.com
Use Case 5: Intent Analysis
Let’s say you have some text … … and a database or spreadsheet with columns
“I want to learn more about
the premium plan you have
for frequent fliers.”
… and you have to fill in the database fields
from the information in the text …
Product Intent Constraint
Premium
plan
Inquire Frequent
fliers
Aiaioo Labs vaklipi.com
Use Case 5: Intent Analysis
Live demo: http://www.aiaioo.com:8080/annotator-0.1/automation/demoView/2
Aiaioo Labs vaklipi.com
Use Case 6: Sales Lead List Extraction
Let’s say you have some text … … and a database or spreadsheet with columns
“Tim Cook is the CEO of Apple a firm in the computer industry…”
… and you have to build a leads list of CEOs by industry …
CEO Firm Industry
Tim Cook Apple Computers
Aiaioo Labs vaklipi.com
Use Case 6:
Sales Function – Sales Lead List Extraction
Video demonstration: http://selasdia.com/lead-finder/
(Selasdia is the brand of a robotic sales assistant developed on the Vaklipi platform)
Aiaioo Labs vaklipi.com
Use Case 6: CXO and Email Finder
Video demonstration: http://selasdia.com/cxo-finder/
(Selasdia is the brand of a robotic sales assistant developed on the Vaklipi platform)
Aiaioo Labs vaklipi.com
Vaklipi Classification Tools
Type = “Tools”
Aiaioo Labs vaklipi.com
b. Vaklipi Classification Tools
Use Cases:
1. Problem Routing (Assignment)
2. Topic Classification
3. Sentiment Classification
4. Subjectivity Classification
5. Intention Classification
6. Language Identification
7. Gender Identification
(3, 4, 5, 6 and 7 are supported out of the box, while
1 and 2 will require customization)
Aiaioo Labs vaklipi.com
Use Case 1: Problem Routing
Let’s say you have some text … … and you want to mark it as one of …
“John Chambers of Springfield, MA
reported a problem with the clutch
on his Ford Ranger purchased in
Boston, MA in 2005.”
… and so route to customer service or sales
Report
Inquiry
Aiaioo Labs vaklipi.com
Use Case 2: Topic Classification
Aiaioo Labs vaklipi.com
Let’s say you have a news item… … and you want to mark it as one of …
“Luxembourg Foreign Minister Jean Asselborn
has called for Hungary to be suspended or
even expelled from the European Union
because of its "massive violation" of EU
fundamental values..”
… and so decide the news category
Politics
Sports
Use Case 3: Sentiment Classification
Aiaioo Labs vaklipi.com
Let’s say you have a review … … and you want to mark it as one of …
“WOW...that's all I can say. I've been waiting for
this one to come out on Blu-ray for quite awhile...
and I can truly say, it was worth the wait. Finally,
this movie looks like it did in the theater...the colors,
details and clarity are amazing...”
… and so decide the news category
Positive
Negative
Use Case 4: Subjectivity Classification
You can use Vaklipi to decide if there is an opinion/sentiment in the
sentence in the first place …
“I am sad that Steve Jobs died.” Subjective
“Steve Jobs died.” Objective
Aiaioo Labs vaklipi.com
Use Case 5: Intent Classification
We provided an intent classifier with Vaklipi …
“Who has a typewriter for sale?”
• Purchase intention
Aiaioo Labs vaklipi.com
Use Case 6: Language Identification
(A very accurate language identifier is provided with Vaklipi) …
Live demo: http://www.aiaioo.com:8080/annotator-0.1/automation/demoView/6
Aiaioo Labs vaklipi.com
Use Case 7: Gender Identification
(A multilingual gender identifier is provided with Vaklipi) …
Male Female
Arun
Ashwin
Akshata
Aparna
Out-of-the-box models available
Aiaioo Labs vaklipi.com
All the above use cases illustrate a
technique called Classification
There are two types of classification supported by the Vaklipi
platform:
1. Classification using Machine Learning
2. Classification using Linguistic Resources (Gazetteers)
Advantages of using the Vaklipi platform:
1. Several pre-built classifiers available out of the box
2. Support for rapid development of new classifiers
Aiaioo Labs vaklipi.com
Vaklipi Exploration Tools
Aiaioo Labs vaklipi.com
c. Vaklipi Exploration Tools
Use Cases:
1. Automatic Label Generation for Grouping / Retrieval
2. Frequent Issue Detection
3. Key Issue Detection
4. Churn Prediction
5. Product Propensity Analysis
6. Shift Detection
7. Summarization
8. Bootstrapping Extraction Tools
(All are supported out of the box – the raw data will need to
be set up for the tool but no training data is needed)
Aiaioo Labs vaklipi.com
Use Case 1: Data Organization
ABSTRACT
A challenging problem faced by researchers and
developers of distributed real-time and embedded (DRE)
systems is devising and implementing effective adaptive
resource management strategies that can meet end-to-
end quality of service (QoS) requirements in varying
operational conditions. This paper presents two
contributions to research in adaptive resource
management for DRE systems.
How do you label/organize the following research abstract?
Aiaioo Labs vaklipi.com
Use Case 1: Data Organization
1. adaptive resource management
2. distributed real-time embedded system
3. end-to-end quality of service
4. service end-to-end quality
5. hybrid adaptive resourcemanagement middleware
6. hybrid control technique
7. real-time video distribution system
8. real-time corba specification
9. video encoding/decoding
10. resource reservation mechanism
11. dynamic environment
These tags that were chosen by
a human.
Aiaioo Labs vaklipi.com
Use Case 1: Data Organization
1. resource utilization
2. base station
3. qos requirements
4. dre multimedia
5. a/v streaming service
6. frame rate
7. end-to-end qos requirements
8. adaptive resource management
9. network bandwidth
10. application qos
11. picture resolution
These tags that were chosen by
a machine.
Aiaioo Labs vaklipi.com
Use Case 2: Frequent Issue Detection
The customer complained about
delays in processing incoming
international payments ...
Textual Events
I helped the customer recover
his lost password after he was
locked out of his account ...
Method:
Take a bunch of textual events
such as these and reduce them
to a set of issues such as:
a) international payment delays
b) account lockout
Aiaioo Labs vaklipi.com
Use Case 3: Key Issue Detection
The customer complained about
delays in processing incoming
international payments ...
Textual Events
I helped the customer recover
his lost password after he was
locked out of his account ...
Method:
Take a bunch of
textual events such
as these that are
correlated with
adverse business
events and reduce
them to a set of
issues such as:
a) payment delay
b) account lockout
Account balance < Rs.
1,00,000
Adverse Business Events
Online account usage
frequency drops to 0 for
3 months.
Aiaioo Labs vaklipi.com
Use Case 4: Churn Analysis
Applications to Churn Analysis
The customer is having a lot
of issues with the online
portal.
Textual Events
The customer was locked out
of his account again ...
Account balance < Rs.
1,00,000
Adverse Business Events
Online account usage
frequency drops to 0 for
3 months.
Aiaioo Labs vaklipi.com
Use Case 5: Purchase Propensity Analysis
Applications to Purchase Propensity Analysis
The customer inquired about
accessing investment banking
services.
Textual Events
Customer requested a credit
line.
New investment made
Propitious Business Events
Increased safe use of
credit line.
Aiaioo Labs vaklipi.com
Use Cases 6: Shift Detection
Shift detection is a form of trend analysis useful for
marketing or planning.
Category Leaders:
2010
1) Marketo
2) Radian6
3) Pardot
4) Alchemy
5) Hubspot
6) Act-On
7) Lexalytics
Category Leaders:
2015
1) Hubspot
2) Marketo
3) Act-On
4) Salesforce
5) IBM
6) Oracle
7) Adobe
Aiaioo Labs vaklipi.com
Use Case 7: Summarizing Feedback
Applications to processing large interaction streams
Aiaioo Labs vaklipi.com
Vaklipi Generation Tools
Aiaioo Labs vaklipi.com
d. Vaklipi Generation Tools
Use Cases:
1. Text Entry Assistance
2. Smart User Interfaces
3. Customer Support Assistance
4. OCR/Speech Output Correction
5. Question Answering
6. Summarization
(These require training data and extensive customization
and are mostly used where operational efficiency is
important)
Aiaioo Labs vaklipi.com
Use Case 1: Text Entry Assistance (Auto-Suggest)
How do you help a text entry operator make fewer mistakes?
Vaklipi allows you to develop models which predict what someone is
about to type.
These can be integrated into user interfaces to create auto-suggest systems.
Aiaioo Labs vaklipi.com
Use Case 2: Smart User Interfaces
How do you guide a user to use a machine learning tool so that
it makes fewer mistakes?
Vaklipi allows you to develop models that can guide usage intelligently.
These require an integration with user interfaces.
Aiaioo Labs vaklipi.com
Use Case 3: Customer Support Assistance
How do you guide a customer support person to answer
questions faster?
Vaklipi allows you to develop answer retrieval and customization systems that
can make the job of a CSR easier.
Aiaioo Labs vaklipi.com
Use Case 4: OCR/Speech Output Correction
How do you correct the output of a machine learning tool so
that it makes fewer mistakes?
Vaklipi allows you to develop models that can correct other
machine learning algorithms.
Source Channel
Decoder
Aiaioo Labs vaklipi.com
Use Case 5:
Answering Questions
First you classify the question into one of 3 types… and these are…
“Do you ship your cars to Boston, MA?”
“Who is the CEO of Apple?”
“Why is the sky blue?”
Yes/No questions
Factoid questions
Non-factoid questions
Aiaioo Labs vaklipi.com
Use Case 5:
Answering Questions
Look for answers in databases that you created using entity / relationship extraction
“Do you ship your cars to Boston, MA?”
“Who is the CEO of Apple?”
“Why is the sky blue?”
Product Ships To
Cars USA
CEO Firm
Tim
Cook
Apple
Aiaioo Labs vaklipi.com
Use Case 5:
Answering Questions (Who is the CEO of Adobe?)
Video Demo: http://selasdia.com/cxo-finder/
Aiaioo Labs vaklipi.com
Use Case 6: Summarization
Useful for processing / distilling large interaction streams
Aiaioo Labs vaklipi.com
THE END
Bangalore, India
http://www.vaklipi.com
Aiaioo Labs vaklipi.com

More Related Content

Similar to Vaklipi Text Analytics Tools

Negotiating with Intellectual Ventures (IV)? What About Their Other Funds?
Negotiating with Intellectual Ventures (IV)? What About Their Other Funds?Negotiating with Intellectual Ventures (IV)? What About Their Other Funds?
Negotiating with Intellectual Ventures (IV)? What About Their Other Funds?Erik Oliver
 
Testing the Intelligence of your AI
Testing the Intelligence of your AITesting the Intelligence of your AI
Testing the Intelligence of your AIIosif Itkin
 
IWE 2480 - An Ecosystem of Innovation: Creating Cognitive Apps Powered by IB...
IWE 2480 - An Ecosystem of Innovation:  Creating Cognitive Apps Powered by IB...IWE 2480 - An Ecosystem of Innovation:  Creating Cognitive Apps Powered by IB...
IWE 2480 - An Ecosystem of Innovation: Creating Cognitive Apps Powered by IB...Carmine DiMascio
 
Oltre l’intelligenza Artificiale: agire alla velocità del pensiero
Oltre l’intelligenza Artificiale: agire alla velocità del pensieroOltre l’intelligenza Artificiale: agire alla velocità del pensiero
Oltre l’intelligenza Artificiale: agire alla velocità del pensieroJürgen Ambrosi
 
How Corporate Culture and Emloyee Engagement Unlock Market Intelligence Poten...
How Corporate Culture and Emloyee Engagement Unlock Market Intelligence Poten...How Corporate Culture and Emloyee Engagement Unlock Market Intelligence Poten...
How Corporate Culture and Emloyee Engagement Unlock Market Intelligence Poten...IntelCollab.com
 
Sharepoint Document Conversion
Sharepoint Document ConversionSharepoint Document Conversion
Sharepoint Document ConversionColin Gardner
 
Why and How to do a Software Startup
Why and How to do a Software StartupWhy and How to do a Software Startup
Why and How to do a Software Startupkyptin
 
Only the resources provided in the links bellow may be used in wri.docx
Only the resources provided in the links bellow may be used in wri.docxOnly the resources provided in the links bellow may be used in wri.docx
Only the resources provided in the links bellow may be used in wri.docxvannagoforth
 
WIPS Global Brochure, New
WIPS Global Brochure, NewWIPS Global Brochure, New
WIPS Global Brochure, Newshikha gupta
 
Splunk for ITOps
Splunk for ITOpsSplunk for ITOps
Splunk for ITOpsSplunk
 
Data Science Application in Business Portfolio & Risk Management
Data Science Application in Business Portfolio & Risk ManagementData Science Application in Business Portfolio & Risk Management
Data Science Application in Business Portfolio & Risk ManagementData Science Thailand
 
Apache Flink: Real-World Use Cases for Streaming Analytics
Apache Flink: Real-World Use Cases for Streaming AnalyticsApache Flink: Real-World Use Cases for Streaming Analytics
Apache Flink: Real-World Use Cases for Streaming AnalyticsSlim Baltagi
 
Lab-4 Reconnaissance and Information Gathering  A hacker.docx
Lab-4 Reconnaissance and Information Gathering         A hacker.docxLab-4 Reconnaissance and Information Gathering         A hacker.docx
Lab-4 Reconnaissance and Information Gathering  A hacker.docxLaticiaGrissomzz
 
Writing Apa Research Paper
Writing Apa Research PaperWriting Apa Research Paper
Writing Apa Research PaperDiana Marquez
 

Similar to Vaklipi Text Analytics Tools (20)

Finding industry reports
Finding industry reportsFinding industry reports
Finding industry reports
 
Finding industry reports
Finding industry reportsFinding industry reports
Finding industry reports
 
Finding industry reports
Finding industry reportsFinding industry reports
Finding industry reports
 
Negotiating with Intellectual Ventures (IV)? What About Their Other Funds?
Negotiating with Intellectual Ventures (IV)? What About Their Other Funds?Negotiating with Intellectual Ventures (IV)? What About Their Other Funds?
Negotiating with Intellectual Ventures (IV)? What About Their Other Funds?
 
Patent database
Patent databasePatent database
Patent database
 
Testing the Intelligence of your AI
Testing the Intelligence of your AITesting the Intelligence of your AI
Testing the Intelligence of your AI
 
IWE 2480 - An Ecosystem of Innovation: Creating Cognitive Apps Powered by IB...
IWE 2480 - An Ecosystem of Innovation:  Creating Cognitive Apps Powered by IB...IWE 2480 - An Ecosystem of Innovation:  Creating Cognitive Apps Powered by IB...
IWE 2480 - An Ecosystem of Innovation: Creating Cognitive Apps Powered by IB...
 
Oltre l’intelligenza Artificiale: agire alla velocità del pensiero
Oltre l’intelligenza Artificiale: agire alla velocità del pensieroOltre l’intelligenza Artificiale: agire alla velocità del pensiero
Oltre l’intelligenza Artificiale: agire alla velocità del pensiero
 
How Corporate Culture and Emloyee Engagement Unlock Market Intelligence Poten...
How Corporate Culture and Emloyee Engagement Unlock Market Intelligence Poten...How Corporate Culture and Emloyee Engagement Unlock Market Intelligence Poten...
How Corporate Culture and Emloyee Engagement Unlock Market Intelligence Poten...
 
Sharepoint Document Conversion
Sharepoint Document ConversionSharepoint Document Conversion
Sharepoint Document Conversion
 
Why and How to do a Software Startup
Why and How to do a Software StartupWhy and How to do a Software Startup
Why and How to do a Software Startup
 
Only the resources provided in the links bellow may be used in wri.docx
Only the resources provided in the links bellow may be used in wri.docxOnly the resources provided in the links bellow may be used in wri.docx
Only the resources provided in the links bellow may be used in wri.docx
 
WIPS Global Brochure, New
WIPS Global Brochure, NewWIPS Global Brochure, New
WIPS Global Brochure, New
 
Splunk for ITOps
Splunk for ITOpsSplunk for ITOps
Splunk for ITOps
 
Data Science Application in Business Portfolio & Risk Management
Data Science Application in Business Portfolio & Risk ManagementData Science Application in Business Portfolio & Risk Management
Data Science Application in Business Portfolio & Risk Management
 
Apache Flink: Real-World Use Cases for Streaming Analytics
Apache Flink: Real-World Use Cases for Streaming AnalyticsApache Flink: Real-World Use Cases for Streaming Analytics
Apache Flink: Real-World Use Cases for Streaming Analytics
 
Ultimate slideshare post
Ultimate slideshare postUltimate slideshare post
Ultimate slideshare post
 
Ultimate slideshare post
Ultimate slideshare postUltimate slideshare post
Ultimate slideshare post
 
Lab-4 Reconnaissance and Information Gathering  A hacker.docx
Lab-4 Reconnaissance and Information Gathering         A hacker.docxLab-4 Reconnaissance and Information Gathering         A hacker.docx
Lab-4 Reconnaissance and Information Gathering  A hacker.docx
 
Writing Apa Research Paper
Writing Apa Research PaperWriting Apa Research Paper
Writing Apa Research Paper
 

More from aiaioo

Document Analysis with Deep Learning
Document Analysis with Deep LearningDocument Analysis with Deep Learning
Document Analysis with Deep Learningaiaioo
 
Deep Learning through Pytorch Exercises
Deep Learning through Pytorch ExercisesDeep Learning through Pytorch Exercises
Deep Learning through Pytorch Exercisesaiaioo
 
Learning Non-Linear Functions for Text Classification
Learning Non-Linear Functions for Text ClassificationLearning Non-Linear Functions for Text Classification
Learning Non-Linear Functions for Text Classificationaiaioo
 
Arduino for Indian Languages
Arduino for Indian LanguagesArduino for Indian Languages
Arduino for Indian Languagesaiaioo
 
Fun with Text - Hacking Text Analytics
Fun with Text - Hacking Text AnalyticsFun with Text - Hacking Text Analytics
Fun with Text - Hacking Text Analyticsaiaioo
 
Vaklipi (Natural Language Programming and Queries)
Vaklipi (Natural Language Programming and Queries)Vaklipi (Natural Language Programming and Queries)
Vaklipi (Natural Language Programming and Queries)aiaioo
 
Statistics for linguistics
Statistics for linguisticsStatistics for linguistics
Statistics for linguisticsaiaioo
 
Rules engines to machine learning
Rules engines to machine learningRules engines to machine learning
Rules engines to machine learningaiaioo
 
Aiaioo labs - Only Slightly Futuristic
Aiaioo labs - Only Slightly FuturisticAiaioo labs - Only Slightly Futuristic
Aiaioo labs - Only Slightly Futuristicaiaioo
 

More from aiaioo (9)

Document Analysis with Deep Learning
Document Analysis with Deep LearningDocument Analysis with Deep Learning
Document Analysis with Deep Learning
 
Deep Learning through Pytorch Exercises
Deep Learning through Pytorch ExercisesDeep Learning through Pytorch Exercises
Deep Learning through Pytorch Exercises
 
Learning Non-Linear Functions for Text Classification
Learning Non-Linear Functions for Text ClassificationLearning Non-Linear Functions for Text Classification
Learning Non-Linear Functions for Text Classification
 
Arduino for Indian Languages
Arduino for Indian LanguagesArduino for Indian Languages
Arduino for Indian Languages
 
Fun with Text - Hacking Text Analytics
Fun with Text - Hacking Text AnalyticsFun with Text - Hacking Text Analytics
Fun with Text - Hacking Text Analytics
 
Vaklipi (Natural Language Programming and Queries)
Vaklipi (Natural Language Programming and Queries)Vaklipi (Natural Language Programming and Queries)
Vaklipi (Natural Language Programming and Queries)
 
Statistics for linguistics
Statistics for linguisticsStatistics for linguistics
Statistics for linguistics
 
Rules engines to machine learning
Rules engines to machine learningRules engines to machine learning
Rules engines to machine learning
 
Aiaioo labs - Only Slightly Futuristic
Aiaioo labs - Only Slightly FuturisticAiaioo labs - Only Slightly Futuristic
Aiaioo labs - Only Slightly Futuristic
 

Recently uploaded

Conf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
Conf42-LLM_Adding Generative AI to Real-Time Streaming PipelinesConf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
Conf42-LLM_Adding Generative AI to Real-Time Streaming PipelinesTimothy Spann
 
Biometric Authentication: The Evolution, Applications, Benefits and Challenge...
Biometric Authentication: The Evolution, Applications, Benefits and Challenge...Biometric Authentication: The Evolution, Applications, Benefits and Challenge...
Biometric Authentication: The Evolution, Applications, Benefits and Challenge...GQ Research
 
Data Analysis Project : Targeting the Right Customers, Presentation on Bank M...
Data Analysis Project : Targeting the Right Customers, Presentation on Bank M...Data Analysis Project : Targeting the Right Customers, Presentation on Bank M...
Data Analysis Project : Targeting the Right Customers, Presentation on Bank M...Boston Institute of Analytics
 
Predicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdfPredicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdfBoston Institute of Analytics
 
RABBIT: A CLI tool for identifying bots based on their GitHub events.
RABBIT: A CLI tool for identifying bots based on their GitHub events.RABBIT: A CLI tool for identifying bots based on their GitHub events.
RABBIT: A CLI tool for identifying bots based on their GitHub events.natarajan8993
 
原版1:1定制南十字星大学毕业证(SCU毕业证)#文凭成绩单#真实留信学历认证永久存档
原版1:1定制南十字星大学毕业证(SCU毕业证)#文凭成绩单#真实留信学历认证永久存档原版1:1定制南十字星大学毕业证(SCU毕业证)#文凭成绩单#真实留信学历认证永久存档
原版1:1定制南十字星大学毕业证(SCU毕业证)#文凭成绩单#真实留信学历认证永久存档208367051
 
20240419 - Measurecamp Amsterdam - SAM.pdf
20240419 - Measurecamp Amsterdam - SAM.pdf20240419 - Measurecamp Amsterdam - SAM.pdf
20240419 - Measurecamp Amsterdam - SAM.pdfHuman37
 
Student Profile Sample report on improving academic performance by uniting gr...
Student Profile Sample report on improving academic performance by uniting gr...Student Profile Sample report on improving academic performance by uniting gr...
Student Profile Sample report on improving academic performance by uniting gr...Seán Kennedy
 
Defining Constituents, Data Vizzes and Telling a Data Story
Defining Constituents, Data Vizzes and Telling a Data StoryDefining Constituents, Data Vizzes and Telling a Data Story
Defining Constituents, Data Vizzes and Telling a Data StoryJeremy Anderson
 
ASML's Taxonomy Adventure by Daniel Canter
ASML's Taxonomy Adventure by Daniel CanterASML's Taxonomy Adventure by Daniel Canter
ASML's Taxonomy Adventure by Daniel Cantervoginip
 
Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024
Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024
Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024thyngster
 
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024Timothy Spann
 
NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...
NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...
NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...Boston Institute of Analytics
 
LLMs, LMMs, their Improvement Suggestions and the Path towards AGI
LLMs, LMMs, their Improvement Suggestions and the Path towards AGILLMs, LMMs, their Improvement Suggestions and the Path towards AGI
LLMs, LMMs, their Improvement Suggestions and the Path towards AGIThomas Poetter
 
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改yuu sss
 
Thiophen Mechanism khhjjjjjjjhhhhhhhhhhh
Thiophen Mechanism khhjjjjjjjhhhhhhhhhhhThiophen Mechanism khhjjjjjjjhhhhhhhhhhh
Thiophen Mechanism khhjjjjjjjhhhhhhhhhhhYasamin16
 
办理学位证中佛罗里达大学毕业证,UCF成绩单原版一比一
办理学位证中佛罗里达大学毕业证,UCF成绩单原版一比一办理学位证中佛罗里达大学毕业证,UCF成绩单原版一比一
办理学位证中佛罗里达大学毕业证,UCF成绩单原版一比一F sss
 
Multiple time frame trading analysis -brianshannon.pdf
Multiple time frame trading analysis -brianshannon.pdfMultiple time frame trading analysis -brianshannon.pdf
Multiple time frame trading analysis -brianshannon.pdfchwongval
 
RadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfRadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfgstagge
 

Recently uploaded (20)

Conf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
Conf42-LLM_Adding Generative AI to Real-Time Streaming PipelinesConf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
Conf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
 
Biometric Authentication: The Evolution, Applications, Benefits and Challenge...
Biometric Authentication: The Evolution, Applications, Benefits and Challenge...Biometric Authentication: The Evolution, Applications, Benefits and Challenge...
Biometric Authentication: The Evolution, Applications, Benefits and Challenge...
 
Data Analysis Project : Targeting the Right Customers, Presentation on Bank M...
Data Analysis Project : Targeting the Right Customers, Presentation on Bank M...Data Analysis Project : Targeting the Right Customers, Presentation on Bank M...
Data Analysis Project : Targeting the Right Customers, Presentation on Bank M...
 
Predicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdfPredicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdf
 
RABBIT: A CLI tool for identifying bots based on their GitHub events.
RABBIT: A CLI tool for identifying bots based on their GitHub events.RABBIT: A CLI tool for identifying bots based on their GitHub events.
RABBIT: A CLI tool for identifying bots based on their GitHub events.
 
原版1:1定制南十字星大学毕业证(SCU毕业证)#文凭成绩单#真实留信学历认证永久存档
原版1:1定制南十字星大学毕业证(SCU毕业证)#文凭成绩单#真实留信学历认证永久存档原版1:1定制南十字星大学毕业证(SCU毕业证)#文凭成绩单#真实留信学历认证永久存档
原版1:1定制南十字星大学毕业证(SCU毕业证)#文凭成绩单#真实留信学历认证永久存档
 
20240419 - Measurecamp Amsterdam - SAM.pdf
20240419 - Measurecamp Amsterdam - SAM.pdf20240419 - Measurecamp Amsterdam - SAM.pdf
20240419 - Measurecamp Amsterdam - SAM.pdf
 
Student Profile Sample report on improving academic performance by uniting gr...
Student Profile Sample report on improving academic performance by uniting gr...Student Profile Sample report on improving academic performance by uniting gr...
Student Profile Sample report on improving academic performance by uniting gr...
 
Defining Constituents, Data Vizzes and Telling a Data Story
Defining Constituents, Data Vizzes and Telling a Data StoryDefining Constituents, Data Vizzes and Telling a Data Story
Defining Constituents, Data Vizzes and Telling a Data Story
 
ASML's Taxonomy Adventure by Daniel Canter
ASML's Taxonomy Adventure by Daniel CanterASML's Taxonomy Adventure by Daniel Canter
ASML's Taxonomy Adventure by Daniel Canter
 
Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024
Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024
Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024
 
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
 
NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...
NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...
NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...
 
LLMs, LMMs, their Improvement Suggestions and the Path towards AGI
LLMs, LMMs, their Improvement Suggestions and the Path towards AGILLMs, LMMs, their Improvement Suggestions and the Path towards AGI
LLMs, LMMs, their Improvement Suggestions and the Path towards AGI
 
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
 
Thiophen Mechanism khhjjjjjjjhhhhhhhhhhh
Thiophen Mechanism khhjjjjjjjhhhhhhhhhhhThiophen Mechanism khhjjjjjjjhhhhhhhhhhh
Thiophen Mechanism khhjjjjjjjhhhhhhhhhhh
 
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
 
办理学位证中佛罗里达大学毕业证,UCF成绩单原版一比一
办理学位证中佛罗里达大学毕业证,UCF成绩单原版一比一办理学位证中佛罗里达大学毕业证,UCF成绩单原版一比一
办理学位证中佛罗里达大学毕业证,UCF成绩单原版一比一
 
Multiple time frame trading analysis -brianshannon.pdf
Multiple time frame trading analysis -brianshannon.pdfMultiple time frame trading analysis -brianshannon.pdf
Multiple time frame trading analysis -brianshannon.pdf
 
RadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfRadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdf
 

Vaklipi Text Analytics Tools

  • 1. Bangalore, India A Platform for Text Analytics Vaklipi http://www.vaklipi.com Aiaioo Labs vaklipi.com
  • 2. Aiaioo Labs • Research Lab – Natural Language Processing – Machine Learning – Information Retrieval • Vaklipi – Vaklipi is the brand under which our text processing tools are distributed. – Vak means Speech or Spoken Language – Lipi means Writing or Written Language Aiaioo Labs vaklipi.com
  • 3. What can Vaklipi help you get out of text? The opportunities vary by vertical: 1. Aerospace / Defense / Automotive – Compliance process automation / Technical specification standardization / Competitive intelligence and customer feedback processing 2. Healthcare / Life sciences – Reporting / Storing, retrieving and analyzing patents and publications / Extraction of actionable insights from research and competitive intelligence 3. Legal and Government – Legal and administrative filings / Case document and administrative record management, e-discovery / Extraction of facts from legal and administrative documents (land records, case files) 4. FMCG – Routine reports, invoicing, tax documents / Sentiment and Intention Analysis / Call center insights 5. BFSI – Reporting / Churn and product propensity analysis / Forms processing, customer care, monitoring, fraud detection Aiaioo Labs vaklipi.com
  • 4. How do we unlock the value in “text”? Output Text / Store and Analyse Text / Ingest Text d. Vaklipi Generation Tools a. Vaklipi Extraction Tools b. Vaklipi Classification Tools c. Vaklipi Exploration Tools Aiaioo Labs vaklipi.com
  • 6. a. Vaklipi Extraction Tools Use Cases: 1. Customer Service 2. Land Records 3. Mergers and Acquisitions 4. Sentiment Analysis 5. Intention Analysis 6. Sales Lead List Extraction (4, 5 and 6 are supported out of the box, while 1, 2 and 3 will require customization) Aiaioo Labs vaklipi.com
  • 7. Use Case 1: Customer Service Let’s say you have some text … … and a database or spreadsheet with columns “John Chambers of Springfield, MA reported a problem with the clutch on his Ford Ranger purchased in Boston, MA in 2005.” … and you have to fill in the database fields from the information in the text … Reporter Location (of Reporter) Product Aiaioo Labs vaklipi.com
  • 8. Use Case 1: Customer Service “John Chambers of Springfield, MA reported a problem with the clutch on his Ford Ranger purchased in Boston, MA in 2005.” Entities are pieces of text that could go into the fields in the database. Relations connect the entities that belong in a row. Reporter Location Product John Chambers Springfield, MA Ford Ranger Aiaioo Labs vaklipi.com
  • 9. Use Case 2: Government – Land Records Let’s say you have some text … … and a database or spreadsheet with columns “Property K45L234 (lot 23-24) in Wake County of 3000 sq ft was sold to James Fischer on 3-30-1997 …” … and you have to fill in the database fields from the information in the text … Title# Lot County Aiaioo Labs vaklipi.com
  • 10. Use Case 3: Legal Domain – Mergers/Acquisitions Let’s say you have some text … … and a database or spreadsheet with columns “Acme Financials, a subsidiary of Lehman Sisters, was acquired by John Doe Corp on 5/26/2001.” … and you have to fill in the database fields from the information in the text … Acquirer Acquired Date Aiaioo Labs vaklipi.com
  • 11. Use Case 4: Sentiment Analysis Let’s say you have some text … … and a database or spreadsheet with columns “The Canon 3F model is sturdy and excellent for studio use.” … and you have to fill in the database fields from the information in the text … Product Sentiment Purpose Canon 3F Positive Studio Use Aiaioo Labs vaklipi.com
  • 12. Use Case 4: Sentiment Analysis Live demo: http://www.aiaioo.com:8080/annotator-0.1/automation/demoView/1 Aiaioo Labs vaklipi.com
  • 13. Use Case 4: Sentiment Analysis is available in many languages Live demo: http://www.aiaioo.com:8080/annotator-0.1/automation/demoView/1 Aiaioo Labs vaklipi.com
  • 14. Use Case 5: Intent Analysis Let’s say you have some text … … and a database or spreadsheet with columns “I want to learn more about the premium plan you have for frequent fliers.” … and you have to fill in the database fields from the information in the text … Product Intent Constraint Premium plan Inquire Frequent fliers Aiaioo Labs vaklipi.com
  • 15. Use Case 5: Intent Analysis Live demo: http://www.aiaioo.com:8080/annotator-0.1/automation/demoView/2 Aiaioo Labs vaklipi.com
  • 16. Use Case 6: Sales Lead List Extraction Let’s say you have some text … … and a database or spreadsheet with columns “Tim Cook is the CEO of Apple a firm in the computer industry…” … and you have to build a leads list of CEOs by industry … CEO Firm Industry Tim Cook Apple Computers Aiaioo Labs vaklipi.com
  • 17. Use Case 6: Sales Function – Sales Lead List Extraction Video demonstration: http://selasdia.com/lead-finder/ (Selasdia is the brand of a robotic sales assistant developed on the Vaklipi platform) Aiaioo Labs vaklipi.com
  • 18. Use Case 6: CXO and Email Finder Video demonstration: http://selasdia.com/cxo-finder/ (Selasdia is the brand of a robotic sales assistant developed on the Vaklipi platform) Aiaioo Labs vaklipi.com
  • 19. Vaklipi Classification Tools Type = “Tools” Aiaioo Labs vaklipi.com
  • 20. b. Vaklipi Classification Tools Use Cases: 1. Problem Routing (Assignment) 2. Topic Classification 3. Sentiment Classification 4. Subjectivity Classification 5. Intention Classification 6. Language Identification 7. Gender Identification (3, 4, 5, 6 and 7 are supported out of the box, while 1 and 2 will require customization) Aiaioo Labs vaklipi.com
  • 21. Use Case 1: Problem Routing Let’s say you have some text … … and you want to mark it as one of … “John Chambers of Springfield, MA reported a problem with the clutch on his Ford Ranger purchased in Boston, MA in 2005.” … and so route to customer service or sales Report Inquiry Aiaioo Labs vaklipi.com
  • 22. Use Case 2: Topic Classification Aiaioo Labs vaklipi.com Let’s say you have a news item… … and you want to mark it as one of … “Luxembourg Foreign Minister Jean Asselborn has called for Hungary to be suspended or even expelled from the European Union because of its "massive violation" of EU fundamental values..” … and so decide the news category Politics Sports
  • 23. Use Case 3: Sentiment Classification Aiaioo Labs vaklipi.com Let’s say you have a review … … and you want to mark it as one of … “WOW...that's all I can say. I've been waiting for this one to come out on Blu-ray for quite awhile... and I can truly say, it was worth the wait. Finally, this movie looks like it did in the theater...the colors, details and clarity are amazing...” … and so decide the news category Positive Negative
  • 24. Use Case 4: Subjectivity Classification You can use Vaklipi to decide if there is an opinion/sentiment in the sentence in the first place … “I am sad that Steve Jobs died.” Subjective “Steve Jobs died.” Objective Aiaioo Labs vaklipi.com
  • 25. Use Case 5: Intent Classification We provided an intent classifier with Vaklipi … “Who has a typewriter for sale?” • Purchase intention Aiaioo Labs vaklipi.com
  • 26. Use Case 6: Language Identification (A very accurate language identifier is provided with Vaklipi) … Live demo: http://www.aiaioo.com:8080/annotator-0.1/automation/demoView/6 Aiaioo Labs vaklipi.com
  • 27. Use Case 7: Gender Identification (A multilingual gender identifier is provided with Vaklipi) … Male Female Arun Ashwin Akshata Aparna Out-of-the-box models available Aiaioo Labs vaklipi.com
  • 28. All the above use cases illustrate a technique called Classification There are two types of classification supported by the Vaklipi platform: 1. Classification using Machine Learning 2. Classification using Linguistic Resources (Gazetteers) Advantages of using the Vaklipi platform: 1. Several pre-built classifiers available out of the box 2. Support for rapid development of new classifiers Aiaioo Labs vaklipi.com
  • 30. c. Vaklipi Exploration Tools Use Cases: 1. Automatic Label Generation for Grouping / Retrieval 2. Frequent Issue Detection 3. Key Issue Detection 4. Churn Prediction 5. Product Propensity Analysis 6. Shift Detection 7. Summarization 8. Bootstrapping Extraction Tools (All are supported out of the box – the raw data will need to be set up for the tool but no training data is needed) Aiaioo Labs vaklipi.com
  • 31. Use Case 1: Data Organization ABSTRACT A challenging problem faced by researchers and developers of distributed real-time and embedded (DRE) systems is devising and implementing effective adaptive resource management strategies that can meet end-to- end quality of service (QoS) requirements in varying operational conditions. This paper presents two contributions to research in adaptive resource management for DRE systems. How do you label/organize the following research abstract? Aiaioo Labs vaklipi.com
  • 32. Use Case 1: Data Organization 1. adaptive resource management 2. distributed real-time embedded system 3. end-to-end quality of service 4. service end-to-end quality 5. hybrid adaptive resourcemanagement middleware 6. hybrid control technique 7. real-time video distribution system 8. real-time corba specification 9. video encoding/decoding 10. resource reservation mechanism 11. dynamic environment These tags that were chosen by a human. Aiaioo Labs vaklipi.com
  • 33. Use Case 1: Data Organization 1. resource utilization 2. base station 3. qos requirements 4. dre multimedia 5. a/v streaming service 6. frame rate 7. end-to-end qos requirements 8. adaptive resource management 9. network bandwidth 10. application qos 11. picture resolution These tags that were chosen by a machine. Aiaioo Labs vaklipi.com
  • 34. Use Case 2: Frequent Issue Detection The customer complained about delays in processing incoming international payments ... Textual Events I helped the customer recover his lost password after he was locked out of his account ... Method: Take a bunch of textual events such as these and reduce them to a set of issues such as: a) international payment delays b) account lockout Aiaioo Labs vaklipi.com
  • 35. Use Case 3: Key Issue Detection The customer complained about delays in processing incoming international payments ... Textual Events I helped the customer recover his lost password after he was locked out of his account ... Method: Take a bunch of textual events such as these that are correlated with adverse business events and reduce them to a set of issues such as: a) payment delay b) account lockout Account balance < Rs. 1,00,000 Adverse Business Events Online account usage frequency drops to 0 for 3 months. Aiaioo Labs vaklipi.com
  • 36. Use Case 4: Churn Analysis Applications to Churn Analysis The customer is having a lot of issues with the online portal. Textual Events The customer was locked out of his account again ... Account balance < Rs. 1,00,000 Adverse Business Events Online account usage frequency drops to 0 for 3 months. Aiaioo Labs vaklipi.com
  • 37. Use Case 5: Purchase Propensity Analysis Applications to Purchase Propensity Analysis The customer inquired about accessing investment banking services. Textual Events Customer requested a credit line. New investment made Propitious Business Events Increased safe use of credit line. Aiaioo Labs vaklipi.com
  • 38. Use Cases 6: Shift Detection Shift detection is a form of trend analysis useful for marketing or planning. Category Leaders: 2010 1) Marketo 2) Radian6 3) Pardot 4) Alchemy 5) Hubspot 6) Act-On 7) Lexalytics Category Leaders: 2015 1) Hubspot 2) Marketo 3) Act-On 4) Salesforce 5) IBM 6) Oracle 7) Adobe Aiaioo Labs vaklipi.com
  • 39. Use Case 7: Summarizing Feedback Applications to processing large interaction streams Aiaioo Labs vaklipi.com
  • 41. d. Vaklipi Generation Tools Use Cases: 1. Text Entry Assistance 2. Smart User Interfaces 3. Customer Support Assistance 4. OCR/Speech Output Correction 5. Question Answering 6. Summarization (These require training data and extensive customization and are mostly used where operational efficiency is important) Aiaioo Labs vaklipi.com
  • 42. Use Case 1: Text Entry Assistance (Auto-Suggest) How do you help a text entry operator make fewer mistakes? Vaklipi allows you to develop models which predict what someone is about to type. These can be integrated into user interfaces to create auto-suggest systems. Aiaioo Labs vaklipi.com
  • 43. Use Case 2: Smart User Interfaces How do you guide a user to use a machine learning tool so that it makes fewer mistakes? Vaklipi allows you to develop models that can guide usage intelligently. These require an integration with user interfaces. Aiaioo Labs vaklipi.com
  • 44. Use Case 3: Customer Support Assistance How do you guide a customer support person to answer questions faster? Vaklipi allows you to develop answer retrieval and customization systems that can make the job of a CSR easier. Aiaioo Labs vaklipi.com
  • 45. Use Case 4: OCR/Speech Output Correction How do you correct the output of a machine learning tool so that it makes fewer mistakes? Vaklipi allows you to develop models that can correct other machine learning algorithms. Source Channel Decoder Aiaioo Labs vaklipi.com
  • 46. Use Case 5: Answering Questions First you classify the question into one of 3 types… and these are… “Do you ship your cars to Boston, MA?” “Who is the CEO of Apple?” “Why is the sky blue?” Yes/No questions Factoid questions Non-factoid questions Aiaioo Labs vaklipi.com
  • 47. Use Case 5: Answering Questions Look for answers in databases that you created using entity / relationship extraction “Do you ship your cars to Boston, MA?” “Who is the CEO of Apple?” “Why is the sky blue?” Product Ships To Cars USA CEO Firm Tim Cook Apple Aiaioo Labs vaklipi.com
  • 48. Use Case 5: Answering Questions (Who is the CEO of Adobe?) Video Demo: http://selasdia.com/cxo-finder/ Aiaioo Labs vaklipi.com
  • 49. Use Case 6: Summarization Useful for processing / distilling large interaction streams Aiaioo Labs vaklipi.com