Entrust Corporation

Gicrm Ai LLC

Peter Relan

Sunil Bhargava

Roger W Faulkner

Amol Kelkar

Nikhil Varghese

Hung Tran

Chandra Khatri

Michael Barrett Stern

Utkarsh Mittal

Ben Feldman

Nachiketa Rajpurohit

Described herein is an Autonomous Conversational AI system, which does not require any human configuration or annotation, and is used to have multi-turn dialogs with a user. A typical Conversational AI system consists of three main models: Natural Language Understanding (NLU), Dialog Manager (DM) and Natural Language Generation (NLG), which requires human provided data and configuration. The system proposed herein leverages novel Conversational AI methods which automatically generates conversational AI configuration from any historical conversation logs. The automatically generated configuration contains Auto-Topics, Auto-Subtopics, Auto-Intents, Auto-Responses and Auto-Flows which are used to automatically train NLU, DM and NLG models. Once these models are trained for given conversation logs, the system can be used to have dialog with any user.

Autonomous conversational AI system without any configuration by a human

Online fraud is reduced by identifying suspicious activities in real time and providing alerting so that interdiction may be performed. Historical customer behavior is used to identify and flag deviations in activity patterns. An HTTP data stream is parsed, intelligently filtered, and key data is extracted in real time. The key data is periodically extracted from network traffic and used to update corresponding summaries stored in a fraud data mart. The data mart is constantly incrementally updated so that the most current historical information is available to a rules engine for real time comparison with new customer data and patterns occurring on the network. Fraud-related business signatures are applied to this data stream and/or a data mart to identify suspicious online transactions. By understanding the customer session, the customer's intended use of the online application is derived and possible fraudulent activities identified.

Fraud protection using business process-based customer intent analysis

A method is provided for monitoring, predicting performance of, and managing Business Operations by the simultaneous, real-time Integration, Normalization and Correlation of direct measurements at the Business Layer and other Layers of Business Operations. The other Layers considered, may include, for example, Application and Infrastructure Layers. The system enables the user to automate sophisticated management tasks by Correlating measurements of activity, performance and availability at all Layers of Business Operations. Significantly, the techniques described herein extend the domain of the Correlations across real-time measurements from all Layers of Business Operations, giving central importance to the measurements in the Business Layer within the Correlations.

Enterprise management system for normalization, integration and correlation of business measurements with application and infrastructure measurements

Growing community of inventors

Atherton, CA, United States of America

Peter Relan