System and method for using semantic and syntactic graphs for utterance classification
View Patent ↗Disclosed herein is a system, method and computer readable medium storing instructions related to semantic and syntactic information in a language understanding system. The method embodiment of the invention is a method for classifying utterances during a natural language dialog between a human and a computing device. The method comprises receiving a user utterance; generating a semantic and syntactic graph associated with the received utterance, extracting all n-grams as features from the generated semantic and syntactic graph and classifying the utterance. Classifying the utterance may be performed any number of ways such as using the extracted n-grams, a syntactic and semantic graphs or writing rules.
1. A method comprising:
receiving, during a natural language dialog between a user and a computing device, an utterance;
generating, via the computing device, a semantic and syntactic graph using the utterance, the semantic and syntactic graph comprising an initial state and a final state associated with the utterance and being generated by adding transitions that encode semantic and syntactic categories of words or word sequences within the utterance;
extracting possible combinations of features from the semantic and syntactic graph, to yield extracted n-grams;
comparing the extracted n-grams to previously classified n-grams, to yield a comparison;
classifying, via the computing device, the utterance as being associated with a call type based on the extracted n-grams and the comparison, wherein the call type is chosen from a predefined set of call types and wherein the call type characterizes a communication, to yield a classified utterance; and
responding, via the computing device and with a computer-generated response, to the user in the natural language dialog based on the classified utterance.
2. The method of claim 1 , wherein the semantic and syntactic graph further comprises lexical information.
3. The method of claim 1 , wherein the semantic and syntactic graph further comprises named entities and semantic role labels.
4. The method of claim 1 , wherein the semantic and syntactic graph further comprises one of speech tags and a syntactic parse of the utterance.
5. The method of claim 1 , wherein generating the semantic and syntactic graph extends a feature set of a classifier that classifies the utterance.
6. The method of claim 1 , wherein generating the semantic and syntactic graph associated with the utterance further comprises adding transitions encoding semantic and syntactic categories of words to a word graph.
7. The method of claim 1 , wherein generating the semantic and syntactic graph associated with the utterance further comprises utilizing one of: a part of speech tags, a syntactic parse of the utterance, and generic named entity tags.
8. A system comprising:
a processor; and
a computer-readable storage medium having instructions stored which, when executed by the processor, cause the processor to perform operations comprising:
receiving, during a natural language dialog between a user and a computing device, an utterance;
generating a semantic and syntactic graph using the utterance, the semantic and syntactic graph comprising an initial state and a final state associated with the utterance and being generated by adding transitions that encode semantic and syntactic categories of words or word sequences within the utterance;
extracting possible combinations of features from the semantic and syntactic graph, to yield extracted n-grams;
comparing the extracted n-grams to previously classified n-grams, to yield a comparison;
classifying the utterance as being associated with a call type based on the extracted n-grams and the comparison, wherein the call type is chosen from a predefined set of call types and wherein the call type characterizes a communication, to yield a classified utterance; and
responding, with a computer-generated response, to the user in the natural language dialog based on the classified utterance.
9. The system of claim 8 , wherein the semantic and syntactic graph further comprises lexical information.
10. The system of claim 8 , wherein the semantic and syntactic graph further comprises named entities and semantic role labels.
11. The system of claim 8 , wherein the semantic and syntactic graph further comprises one of speech tags and a syntactic parse of the utterance.
12. The system of claim 8 , wherein generating the semantic and syntactic graph extends a feature set of a classifier that performs classifying the utterance.
13. The system of claim 8 , wherein generating the semantic and syntactic graph associated with the utterance further comprises adding transitions encoding semantic and syntactic categories of words to a word graph.
14. The system of claim 8 , wherein generating the semantic and syntactic graph associated with the utterance further comprises utilizing one of: a part of speech tags, a syntactic parse of the utterance, and generic named entity tags.
15. A computer-readable storage device having instructions stored which, when executed by a computing device, cause the computing device to perform operations comprising:
receiving, during a natural language dialog between a user and the computing device, an utterance;
generating a semantic and syntactic graph using the utterance, the semantic and syntactic graph comprising an initial state and a final state associated with the utterance and being generated by adding transitions that encode semantic and syntactic categories of words or word sequences within the utterance;
extracting possible combinations of features from the semantic and syntactic graph, to yield extracted n-grams;
comparing the extracted n-grams to previously classified n-grams, to yield a comparison;
classifying the utterance as being associated with a call type based on the extracted n-grams and the comparison, wherein the call type is chosen from a predefined set of call types and wherein the call type characterizes a communication, to yield a classified utterance; and
responding, with a computer-generated response, to the user in the natural language dialog based on the classified utterance.
16. The computer-readable storage device of claim 15 , wherein the semantic and syntactic graph further comprises lexical information.
17. The computer-readable storage device of claim 15 , wherein the semantic and syntactic graph further comprises named entities and semantic role labels.
18. The computer-readable storage device of claim 15 , wherein the semantic and syntactic graph further comprises one of speech tags and a syntactic parse of the utterance.
19. The computer-readable storage device of claim 15 , wherein generating the semantic and syntactic graph extends a feature set of a classifier that performs classifying the utterance.
20. The computer-readable storage device of claim 15 , wherein generating the semantic and syntactic graph associated with the utterance further comprises adding transitions encoding semantic and syntactic categories of words to a word graph.