Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Automating Enterprise Business Processes
using AI
Naghi Prasad
Xu Miao
22
AI-driven enterprise applications
•Business processes mapped to an AI engine to enable
business efficiencies.
•4 busine...
33
But then what is AI? – Lessons Learned
•AI is a rich source of interesting tools
Lot more than Deep Learning, CNN, Gen...
4
Neva
AI-Driven Automation for Service and Support
55
Why Neva?
Customer service organizations must improve
support quality while reducing delivery costs.
Key challenges
 F...
77
Structured
Complicated Data Environment
Structured
Operational DBs
Datawarehouse
APs
Enterprise APIs
Structured
Unstruc...
8
Model Driven Intelligent Data Process
Search GraphMySQL Redis
Hive/Pre
sto
Active Learning
Input Query analysis
Knowledg...
www.swooptalent.com
Talent Data Cloud with SwoopTalent
www.swooptalent.com
Your PRIVATE data
backbone
Data from ALL sources
matched & made available
Private Talent Data Cloud
Pr...
www.swooptalent.com
Candidate Profiles on SwoopTalent
External (public) records
Combined data: rich, fresh,
searchable, an...
www.swooptalent.com
Swoop AI Layer
More Structured Less Structured
ATS, CRM
XML, Excel, Flat
Files,
Social Media, Niche
Fo...
Up IQ
Autonomous MarketingTM
Copyright 2016 Confidential & Proprietary | Not Meant for Distribution
Automatically generates statistically relevant
marketing content that is highly personalized
10x better conversion rates
f...
Enterprises Journey to Autonomous Marketing
Data
Sync
Banks Data
Social Data
Public Records
Data
Cleansing
Data is Enginee...
Up IQ to Power Banks: SEM Campaigns, & Landing Pages
Customers Journey, from Discovery to Acquisition
Personalized
Banks
Retail Banks
Mortgage Banks
Online Lenders
Relevant
Ba...
Deep Forest Media
a Rakuten Company
Cross Device Graph
1919
Machine learning models device graph
relationships : naive Bayes modeling &
heuristics for pruning...
Bid Price Optimization
20
• A dynamic pricing algorithm
– maximizes the expected value of gain after winning an auction, o...
2121
But then what is AI? – Lessons Learned
•AI is a rich source of tools
Deep Learning, CNN, Generative Adversarial Netw...
Questions?
Naghi Prasad
Xu Miao
Neva.ai
Neva: Xu Miao xu@neva.ai, Naghi Prasad naghi@neva.ai
UpIQ : Maksym Bychkov, max@up...
Naghi Prasad at AI Frontiers: Building AI systems to automate enterprise process flows
Upcoming SlideShare
Loading in …5
×

Naghi Prasad at AI Frontiers: Building AI systems to automate enterprise process flows

1,237 views

Published on

In this talk we will discuss our experience building AI systems for enterprise process automation. Using examples of real-life deployed AI systems in AdTech, customer service, mortgage financing and recruiting, we discuss our learnings and insights gleaned.

Published in: Technology
  • Be the first to comment

Naghi Prasad at AI Frontiers: Building AI systems to automate enterprise process flows

  1. 1. Automating Enterprise Business Processes using AI Naghi Prasad Xu Miao
  2. 2. 22 AI-driven enterprise applications •Business processes mapped to an AI engine to enable business efficiencies. •4 business processes being automated by AI Customer Support Recruiting  Content Marketing AdTech • We will conclude with lessons learned from being very involved in these companies since inception.
  3. 3. 33 But then what is AI? – Lessons Learned •AI is a rich source of interesting tools Lot more than Deep Learning, CNN, Generative Adversarial Networks!! Suite of techniques to evoke intelligence :  Categorizers, Regression, NLP, Case-Based reasoning etc. •Domain driven rather than technique driven Let the domain drive the problem solving and which techniques you use from the bag •Interesting Data strategies •AI application is like a raisin bread : it is still 90% bread
  4. 4. 4 Neva AI-Driven Automation for Service and Support
  5. 5. 55 Why Neva? Customer service organizations must improve support quality while reducing delivery costs. Key challenges  Fragmented knowledge from disparate knowledge sources and enterprise systems, and decentralized change management.  Inefficient decision-making due to gap between front and back office, frequent changes, and inability to continuously train human agents.  Fractured user experience due to omni-channel, modern support outside work and inefficient, human-based support at work
  6. 6. 77 Structured Complicated Data Environment Structured Operational DBs Datawarehouse APs Enterprise APIs Structured Unstructured Knowledge Articles Forum Posts Screen shots Structured Semi-structured E-mails Log/message history
  7. 7. 8 Model Driven Intelligent Data Process Search GraphMySQL Redis Hive/Pre sto Active Learning Input Query analysis Knowledge Relevance Indexing Ranking BusinessLogic Document understanding output Inference Learning SQL OLTP OLAP SkitLearn
  8. 8. www.swooptalent.com Talent Data Cloud with SwoopTalent
  9. 9. www.swooptalent.com Your PRIVATE data backbone Data from ALL sources matched & made available Private Talent Data Cloud Production ATS - cloud Data from prior ATS CRM and other live systems - cloud Hundreds of millions of social talent records gathered by Swoop Resumes, Spreadsheets, etc
  10. 10. www.swooptalent.com Candidate Profiles on SwoopTalent External (public) records Combined data: rich, fresh, searchable, analyzableInternal ATS records
  11. 11. www.swooptalent.com Swoop AI Layer More Structured Less Structured ATS, CRM XML, Excel, Flat Files, Social Media, Niche Forums, Society Boards Docs, PDF, JPG, Supervised Learning Tokenization Part of Speech Named Entity Recognition Custom Pipelines Unsupervised Learning Clustering Similarity Latent Semantic Analysis, SVD, Word2Vec Topic Modeling (160 Million Profiles) Data Data Data Data Search Semantic Query Processing Topic Modeling Application
  12. 12. Up IQ Autonomous MarketingTM Copyright 2016 Confidential & Proprietary | Not Meant for Distribution
  13. 13. Automatically generates statistically relevant marketing content that is highly personalized 10x better conversion rates for organic search
  14. 14. Enterprises Journey to Autonomous Marketing Data Sync Banks Data Social Data Public Records Data Cleansing Data is Engineered Content Creation Search Content Social Content Email & Text Machine Learning NLPK Data Science • Markovian Modeling
  15. 15. Up IQ to Power Banks: SEM Campaigns, & Landing Pages
  16. 16. Customers Journey, from Discovery to Acquisition Personalized Banks Retail Banks Mortgage Banks Online Lenders Relevant Bank Staff Ranked Bank Staff Content Discovery Search Content Social Content Email & Text • Information Theoretic Scoring • Sentiment Analysis
  17. 17. Deep Forest Media a Rakuten Company
  18. 18. Cross Device Graph 1919 Machine learning models device graph relationships : naive Bayes modeling & heuristics for pruning. 172.0.0.217 Feature engineering (UID, IP, user agent, referral url, login email etc.) Data Collection (cookie-sync, exchanges, ad impression, native sdk, 3rd party data Identify users across smartphones, tablets & desktops
  19. 19. Bid Price Optimization 20 • A dynamic pricing algorithm – maximizes the expected value of gain after winning an auction, or 𝑏 = 𝑎𝑟𝑔𝑚𝑎𝑥 𝑏 𝐸 𝑔𝑎𝑖𝑛 – adjusts automatically to meet business requirements (ex. CPM margin) using a feedback loop auction data user data win rate win price purchase prediction ctr bidding strategy bid price business requirements alpha • Machine learning models win rate – binary classification (Random forest) win price – regression purchase prediction – binary classification (Random forest) CTR – binary classification
  20. 20. 2121 But then what is AI? – Lessons Learned •AI is a rich source of tools Deep Learning, CNN, Generative Adversarial Networks Categorizers, Regression, NLP, Case-Based reasoning etc. •Domain driven rather than technique driven •Interesting Data strategies •AI application is like a raisin bread : it is still 90% bread
  21. 21. Questions? Naghi Prasad Xu Miao Neva.ai Neva: Xu Miao xu@neva.ai, Naghi Prasad naghi@neva.ai UpIQ : Maksym Bychkov, max@upiq.ai SwoopTalent : Satish Sallakonda satish@swooptalent.com Rakuten : Baiji He, baiji@deepforestmedia.com

×