Upgrade to Pro — share decks privately, control downloads, hide ads and more …

OpenTalks.AI - Влад Беляев, Чат-бот Валера. Как...

OpenTalks.AI
February 15, 2019

OpenTalks.AI - Влад Беляев, Чат-бот Валера. Как автоматизировать поддержку сотрудников и клиентов банка на базе платформы AutoFAQ.ai

OpenTalks.AI

February 15, 2019

Transcript

7 slides
  1. AutoFAQ “A system that gets it” or chatbot “Valera” Use

    case for customer and employee support WWW.AUTOFAQ.AI
  2. WWW.AUTOFAQ.AI AutoFAQ.ai case study “A system that gets it” WWW.AUTOFAQ.AI

    2 pilots, on-premise installation, full stack AI Production launch in August 2018 4 departments – IT SD, HR, Client and Employee support 30% of support cases solved by AutoFAQ Aiming at 50-60% cases
  3. WWW.AUTOFAQ.AI 4 Problem Employee support Can I take money instead

    of a vacation? 70% Do I have to go to a vacation? 61% How can I plan an additional vacation? 82% Compensation for a vacation IT Service Desk Support Retail branches Internal portal Accountants HR
  4. WWW.AUTOFAQ.AI Common chatbot practice “Hype” attitude - We can do

    it / Everyone can do it - Chatbot is a “make it and forget about it” thing - General AI expectations - Data / No Data – who cares “Right” attitude - Work with professionals - A self-learning solution autonomous from a vendor - Keep an eye on the knowledge base - Handle expectations of decision makers - Start with one department, expand to others later
  5. WWW.AUTOFAQ.AI Results 1st line 2nd line 30%* solved by AutoFAQ

    3rd line * Together with the bank we work and aim at 50-60% with new scenarios, integrations and knowledge base improvements Different use case 1st line 2nd line 3rd line 90% correctly classified by AI Needs a lot of data

More Decks by OpenTalks.AI

Other Decks in Science