Enterprise-Grade Chat to be Delivered Quickly

Across the globe organizations are facing higher demand to deliver an intelligent self-service chat experiences. It will be a key challenge to build dynamic and transactional chatbots which can satisfy existing enterprise systems. We would need a pre-built platform which can deliver chatbots

Our Approach
Citta Chat—A Cognitive Chatbot with continuous Learning

Citta Chat is an innovative AI based channel powered by the Citta Platform for creating and deploying purpose-built chatbots for transactional use cases within a short time.

Intelligence Flow

1. Understand
Citta Chat Natural Language Engine supports dynamic training on top of existing enterprise data—like product info, company or contact names. Then it detects these entities, extracts the user intent and passes them to the Cognitive Flow.

2. Process
Citta Chat analyzes the current conversation, long-term bot memory and the goals defined by User. Then it dynamically generates the conversation flow based on each user input.

3. Deliver
Citta Chat requests needed data from internal systems like CRM, Microsoft SharePoint, Zoho or any API. Then it displays the response in the user's channel of choice.

  • Conversational UI
  • Channels
  • Natural Language Analysis
  • Conversational Intelligence
  • Memory
  • Third Party Integrations

Citta Chat provides intuitive, UI components for conversational interaction that makes it easy to select dates, times, forms, items images etc...

  • Immediate response

  • Time capture

  • Login Forms

  • Date capture

  • Image capture

  • Multiple Selections

Citta Chat supports multiple channels

  • Skype

  • Facebook

  • iOS

  • Android

  • Websites

  • Slack

  • Facebook messenger

Citta Chat supports Natural Language understanding from various sources including from the existing enterprise data through web services from Salesforce, CRM, Email, Share Point, Zendesk or any other enterprise product.

  • Entity

  • Intent

  • Synonyms

  • Keywords

  • Text mining

The actual conversation flow is controlled by the cognitive algorithms of Citta Chat, which produce natural conversations with users.

  • Goal Conversations

  • Support Conversations

  • Small Talks

  • Validations

  • Ambiguity

  • Acknowledgements

  • Confirmations

    Past Behaviors
  • Value Suggestions

  • Conversation Predictions

Citta Chat securely stores each interaction with a user in its short- and long-term memory. That can be used to suggest values in subsequent conversations or even to skip some steps.

  • Long-term memory

  • Conversation History

  • Short-term memory

Citta Chat provides flexible REST API integration and samples on how to connect to your existing software infrastructure to deliver personalized user experiences.

  • CRM

  • Share point

  • Sales Force

  • Nintex

  • Microsoft dynamic 365

  • Zoho

Use Cases

Fraud & Risk Management in Government Agencies

Public Sector organizations can better combat fraudulent claims by using Machine Learning to analyze behaviors to identify emerging trends in fraud and abuse so that agencies can act before they cause significant damage

Through the use of machine learning, is able to analyze case files and interview transcripts to identify specific attributes associated with high-risk groups and monitor for suspicious transactions using these criteria

Inspections and Maintenance uses video analytics and machine learning to collect data from remote sensors to model and predict risks, such as mechanical failures, potential breakdowns or public safety threats

This saves the investigator time by pinpointing higher likelihood cases, and in that way, the investigator can take preemptive action

This results in safer working conditions, higher production levels, lower operating costs and more effective incident investigations.

Avoiding Claims Payout Leakages using and RPA

Insurance agents upload images associated with the claims such as those of a damaged car, and an estimate of how much they think the client should receive as a payout based on the damage in the image compares the uploaded image to a backlog of images of various severities of damage and the payouts associated with them then triggers an RPA Bot that checks in the core system if the insurance agent’s payout estimate is more than the payout that other clients received for similarly damaged vehicles

Agent can then decrease the payout that they intend to provide the client based on the payouts that past clients received for similarly damaged vehicles

Competency Assessment & Targeted Recruiting in Government Agencies

Government Agencies can streamline the new hire process using machine learning to extract and understand behavioral attribute patterns from a candidate’s digital interview through audio, video, and text analysis helps identify applicants with the right skills and behavioral attributes for future job success, reducing the recruiting process cycle time