Natural Language Query
First Claim
1. A system and method of natural language data query that eithera) successfully understands all phrases of a user'"'"'s natural language request and automatically converts the request into a database query and returns the answer to the user, orb) alternatively captures unrecognized request phrases and provides a means of “
- teaching the system”
a semantically equivalent user natural language request which will be understood by the system in the future.
0 Assignments
0 Petitions
Accused Products
Abstract
The invention allows end users to obtain immediate, accurate information from structured databases without writing complex database query commands. The invention allows two different, but synchronized, methods of end user information requests: Spoken or typed Natural Language requests, and a Visual Request Specification method. Furthermore, the invention provides a user the means of “teaching the system the correct interpretation” when an information request was misunderstood by the invention'"'"'s Natural Language Understanding module.
116 Citations
11 Claims
-
1. A system and method of natural language data query that either
a) successfully understands all phrases of a user'"'"'s natural language request and automatically converts the request into a database query and returns the answer to the user, or b) alternatively captures unrecognized request phrases and provides a means of “ - teaching the system”
a semantically equivalent user natural language request which will be understood by the system in the future. - View Dependent Claims (2, 3, 6, 7, 8, 9, 10, 11)
- teaching the system”
-
4. An independent “
- setup”
process of capturing the metadata of a database, comprising the elements of;a) capturing and conceptualizing the target data source metadata as a separate “
target concept model”
,b) defining the metadata of the target concept model as entities mapped directly or indirectly to tables and of attributes mapped directly or indirectly to table columns, c) setting metadata-driven “
navigation rules”
for Query Generation,d) setting “
entity restriction”
rules to facilitate NLU disambiguation,e) mapping target concept model entities and attributes to “
parent”
entities and attributes in an ontology model representing a “
target subject area”
of which the database target concept model is an instance, the ontology model being uncoupled from but mapped by inference to any particular target databases within the scope of that subject area.
- setup”
-
5. An independent method creating and organizing a “
- semantic metadata repository”
of a type hierarchy of semantic phrases, independent of concept objects, with the capability toa) map “
concept-centric”
semantic phrases to subject area entities or attributes in a subject area ontology model, thus enriching each subject area ontology as a “
synchronized”
set of both semantic phrases and their mapped concept objects,b) allow concept-centric semantic phrases to refer to entities or attributes in different Subject Area Ontologies, wherein a Subject Concept in one subject area ontology may be mapped to a Predicate Concept in another subject area ontology, provided that the two ontologies are in the same type hierarchy chain of “
more general”
/“
more-specific”
ontologies.
- semantic metadata repository”
Specification