Omnichannel, intelligent, proactive virtual agent
An omni-channel, intelligent, proactive virtual agent system and method of use are provided by which a user may engage in a conversation with the agent to interact with structured and unstructured data of an enterprise that is stored in a domain-specific world model for the enterprise.
1. A system comprising:
a world model comprising a knowledge graph, the knowledge graph representing a portion of enterprise knowledge;
an artificial intelligence agent system that uses the world model to facilitate responding to a query with enterprise domain-directed dialog by defining where in an enterprise system relevant data to the query is found;
an enterprise system in communication with the artificial intelligence agent system and for providing enterprise specific information for the enterprise domain-directed dialog associated with the query being processed by the artificial intelligence agent system; and
a client system that facilitates human interaction with the artificial intelligence agent system with a speech input.
2. The system of claim 1 , wherein the artificial intelligence agent system is deployed on at least one of the enterprise system, the client system, or a cloud-based system.
3. The system of claim 1 , wherein the artificial intelligence agent system is deployed on the client system and interacts with at least one of the enterprise system or the world model via a network.
4. The system of claim 1 , wherein the artificial intelligence agent system is deployed with the enterprise system on one or more enterprise servers for direct interaction between the artificial intelligence agent system and resources of the enterprise system and interacts with the client system via a network.
5. The system of claim 1 , wherein the client system comprises an item of equipment that performs a non-communication function selected from a list of functions consisting of sensing, security, and home automation.
6. The system of claim 1 , wherein the system further comprises a set of application programming interfaces (APIs) that facilitate interfacing the artificial intelligence agent system with the client system.
7. The system of claim 6 , wherein the set of APIs includes a query API that facilitates coordination of the world model and enterprise system resources to at least one of prepare or obtain information for responding to the query.
8. The system of claim 6 , wherein the set of APIs includes a cloud-based call initiating API that facilitates real-time, event-based notification of users of the enterprise system that is coupled to the artificial intelligence agent system.
9. The system of claim 1 , wherein the artificial intelligence agent system comprises a platform portion that operates the world model and an integration engine portion that facilitates interaction between the platform portion and enterprise specific information sources of the enterprise system to respond to queries about the enterprise specific information from the enterprise specific information sources.
10. The system of claim 9 , wherein the integration engine is deployed with the enterprise system.
11. The system of claim 9 , wherein the integration engine converts the enterprise specific information from the enterprise specific information sources to knowledge and stores the knowledge in the world model.
12. The system of claim 9 , wherein the platform portion accesses portions of the world model, thereby activating the integration engine to provide access to a corresponding source of the enterprise specific information.
13. The system of claim 1 , wherein elements in the knowledge graph are linked based on logical and semantic relationships.
14. The system of claim 13 , wherein the knowledge graph facilitates mapping an intent of the query with the enterprise specific information.
15. The system of claim 13 , wherein the knowledge graph facilitates mapping information accessible in the enterprise system with at least one fact accessible in the world model that is relevant to an intent of the query.
16. The system of claim 1 , wherein the world model facilitates semantic understanding of uses of words and terms of an enterprise.
17. The system of claim 1 , wherein the world model is a semantic model that facilitates the artificial intelligence agent system in responding to the query based on semantic relationships of the enterprise specific information representing a portion of enterprise knowledge.
18. The system of claim 1 , wherein the world model comprises information on how to interpret data from the enterprise system.
19. The system of claim 1 , wherein the world model is a graph model comprising a plurality of groups, subgroups, and nodes.
20. The system of claim 1 , wherein the world model comprises a formal ontology and wherein the artificial intelligence agent system uses the world model ontology to match a concept from the query to terminology that has been assigned to the concept.
21. The system of claim 1 , wherein the world model is organized to include at least one quadrant selected from group consisting of internal data, external data, internal knowledge and external knowledge quadrants.
22. The system of claim 1 , wherein the knowledge graph comprises a plurality of layers of knowledge that represents a portion of enterprise knowledge of an enterprise such that elements in the knowledge graph are linkable based on semantic relationships, the plurality of layers comprising:
a global ontology layer of industry-specific knowledge;
a company ontology layer of company-specific knowledge; and
a content layer including links to external data.
23. A computer-implemented method comprising:
receiving a query and facilitating human interaction by enabling dialog with a natural language speech processing technology;
utilizing a world model to facilitate responding to the query by defining where in an enterprise system relevant data to the query is found, wherein the world model ties various types of data to each other and comprises at least one knowledge graph representing a portion of enterprise knowledge;
enabling enterprise domain-directed dialog using the natural language speech processing technology;
providing enterprise specific information for the enterprise domain-directed dialog associated with the query; and
processing the query.
24. A system to engage in an enterprise domain-directed dialog with a user comprising:
a world model comprising a knowledge graph, the knowledge graph representing a portion of enterprise knowledge with nodes that reference a set of semantic topics and links that represent relationships between the set of semantic topics, wherein the world model defines a location of relevant data responsive to a query;
an artificial intelligence agent system that utilizes (a) the world model and (b) a natural language processing system, to process the query, generate a response to the query, and engage in the enterprise domain directed dialog with the user;
an enterprise system in communication with the artificial intelligence agent system, the enterprise system providing enterprise specific information associated with the query for the enterprise domain-directed dialog; and
a client system that receives speech input from the user and interfaces with the artificial intelligence agent system, wherein the speech input comprises the query.