Intelligent Process Automation with Jetson-TX 2 Murali Kaundinya - - PowerPoint PPT Presentation

▶
intelligent process automation with jetson tx 2
SMART_READER_LITE
LIVE PREVIEW

Intelligent Process Automation with Jetson-TX 2 Murali Kaundinya - - PowerPoint PPT Presentation

Oct 14, 2023 47 likes •167 views

Intelligent Process Automation with Jetson-TX 2 Murali Kaundinya Imagine if we could engage enable end users processes to learn to reflect on a workflow or UX from actionable, real-time that can be improved insights from operational

slide-1
SLIDE 1

Intelligent Process Automation with Jetson-TX 2

Murali Kaundinya

slide-2
SLIDE 2

Imagine if we could …

engage end users

to reflect on a workflow or UX that can be improved

  • and then quickly improve it

themselves or recommend and deliver a proposed solution.

  • No different than your experience

participating in OSS project

enable processes to learn

from actionable, real-time insights from operational intelligence

  • augment human decision making

with AI and analytics

  • to experiment faster with dynamic &

intelligent workflows

2

slide-3
SLIDE 3

About me…

  • Murali Kaundinya
  • With Merck & Co. for 2+ Years
  • Horizon 3 - Research & Development
  • Artificial Intelligence
  • Data Science
  • Information Management
  • Previously at :
  • Goldman Sachs
  • UnitedHealth Group (Now Optum)
  • Sun Microsystems (Now Oracle)
  • Started career at
  • NASA Goddard Space Flight

Center, Greenbelt, MD

3

slide-4
SLIDE 4

IPA

Analytics Predictive Prescriptive Augmented Unattended Workflow User Experience NLPs/VPAs Sentiment Analysis Computer Vision Process Invoice Matching Email Analysis Patterns of Work Machine Learning

slide-5
SLIDE 5

Use Cases

Optical Character Recognition Release Engineering Virtual Customer Assistants Data Unification

5

slide-6
SLIDE 6

Architecture

6

Progressive Web Application Embedded BPM Engine Pluggable ML Layer (TensorFlow, Tesseract) 1 WF 2 WF 3

Post Proc Pre

4 5 6 Augmented Analytics

slide-7
SLIDE 7

7

OCR/Document Processing

  • Claims Processing
  • Invoice Processing
  • Language Translation
  • Data Leakage Protection
  • Reduce Errors/Costs

Continuous Delivery

  • Release Notes Generation
  • Change Request
  • CAB Artifacts
  • ML on Impact Analysis
  • Augmented Analytics
  • Approval Workflow

VCAs (aka Chatbots)

  • SOPs/Customer Service
  • Chat Log History
  • Extracting Intent
  • SVMs, Naïve Bayes, LSTMs,

FFNNs

  • NLP w/ Semantic Ontologies
  • Dynamic workflow
slide-8
SLIDE 8

Data Unification

  • 11 source databases
  • Data refreshes
  • OMOP Common Data Model
  • Schema Updates
  • N X M Data Conversions
  • Data Cataloguing algos/tools
  • End-to-End Integration
  • Selective SME Involvement

8 Source: http://OHDSI.org/CDM

Claims EMR Common Data Model Domain Specific Views

Step 1 Step 2 Step 3

Analytics Data Visualization RCT OMICs

Data Similarity Algorithms

Attribute Identification

  • Trigram cosine similarity.
  • TF-IDF cosine similarity.
  • Minimum Desc. Length(Jaccard)
  • Welch’s t-test for numerical values.

Entity Consolidation

  • Data De-duplication
  • Clustering
slide-9
SLIDE 9

DataWrangler

9

FEBRL HoloClean Snorkel

slide-10
SLIDE 10

Target State Architecture

10

Claims EMR Common Data Model Domain Specific Views

Step 1 Step 2 Step 3

Analytics Data Visualization RCT OMICs

slide-11
SLIDE 11

What’s next?

  • End-to-End integration
  • Can GPU databases speed up data unification?
  • Brytlyt
  • MapD, Kinetica, Sqream
  • BlazingDB
  • Blazegraph – Graph Analytics
  • Better data visualization with GPUs
  • MapD

11

slide-12
SLIDE 12

Open architecture that can

engage end users

  • to reflect on a workflow
  • r UX that can be

improved

  • and then quickly

improve it themselves

  • r recommend and

deliver a proposed solution.

  • No different than your

experience participating in OSS project enable processes to learn

  • from actionable, real-time

insights from operational intelligence

  • augment human

decision making with AI and analytics

  • to experiment faster

with dynamic & intelligent workflows

12