Enabling autonomous agents to discriminate between questions and requests
AssigneeOracle International Corporation
InventorsBoris Galitsky, Vishal Vishnoi, Anfernee Xu
Life of this patent
Estimating the term…
Abstract
Systems, devices, and methods of the present invention relate to text classification. A text classification system accesses an utterance of text. The utterance includes at least one word. The text classification system generates a parse tree for the utterance. The parse tree includes at least one terminal node with a word type. The terminal node represents a word of the utterance. The text classification system applies one or more rules to the text. The text classification system then classifies the utterance as a question or a request for an autonomous agent to perform an action.
Claims21
A system for classifying text as a question or a request, comprising:
- a linguistic engine configured to:
- access an utterance comprising at least one word;
- generate a parse tree for the utterance, wherein the parse tree comprises at least one terminal node comprising a word type, wherein each terminal node represents a word or phrase of the utterance;
- and
- apply, to the parse tree, one or more linguistic templates comprising one or more word types;
- and
- a rule engine configured to generate one or more match results obtained from the linguistic engine and determine a classification of the text by applying rules comprising:
- (i) responsive to determining a linguistic template match, classifying the utterance as a request,
- (ii) responsive to determining that the utterance comprises an imperative verb as a first word of the utterance, classifying the utterance as a request,
- (iii) responsive to identifying, in the utterance, one or more predefined request keywords, classifying the utterance as a request, and
- (iv) responsive to identifying, in the utterance, one or more predefined question keywords, classifying the utterance as a question,
- wherein the system is further configured, based on the classification, to send a message to a user device or adjust a configuration of an external device.
A method of using a classification model to classify utterances as questions or requests, the method comprising:
- accessing an utterance of text, wherein the utterance comprises at least one word;
- generating a parse tree for the utterance, wherein the parse tree comprises least one terminal node comprising a word type, wherein each terminal node represents a word or phrase of the utterance;
- determining a classification of the utterance by applying a classification model to the parse tree, the applying comprising:
- determining, for the parse tree, a first similarity score indicating a first match between the utterance and training classes identified as questions, and
- determining, for the parse tree, a second similarity score indicating a second match between the utterance and training classes identified as requests;
- receiving, from the classification model, a classification of the utterance, wherein the classification of the utterance is (i) based on a similarity score for the determined class being higher than other similarity scores or is (ii) unknown.
A method of classifying an utterance as a question or a request, the method comprising:
- accessing an utterance of text, wherein the utterance comprises at least one word;
- generating a parse tree for the utterance, wherein the parse tree comprises at least one terminal node comprising a word type, wherein each terminal node represents a word of the utterance;
- and
- performing one or more operations from a set of operations comprising:
- (i) responsive to determining that a first terminal node of the parse tree comprises an imperative verb, classifying the utterance as a request, and
- (ii) responsive to classifying, in the parse tree, one or more predefined question keywords, classifying the utterance as a question.
(canceled)
Dependencies are inferred from the claim text.