System and method of generating responses to text-based messages

• US 8,566,096 B2
• Filed: 10/10/2012
• Issued: 10/22/2013
• Est. Priority Date: 09/18/2007
• Status: Active Grant

• Current Assignee: Interactions, LLC
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Bangalore, Srinivas, Gilbert, Mazin, Gupta, Narendra
• Primary Examiner(s): Abebe, Daniel D
• Application Number:
  US13/648,647
• Publication Number:
  US 20130035932A1
• Time in Patent Office: 377 Days
• Field of Search: 704/251, 704/257, 704/260, 709/223
• US Class Current: 704/251
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

A system to generate a response to a text-based natural language message includes a user interface, processing device, and a computer-readable storage medium storing executable instructions to generate the response to the text-based natural language message. The instructions and a method for generating the response include identifying a sentence in the text-based natural language message, identifying an input clause in the sentence, and parsing the input clause, thereby defining a relationship between words in the input clause. The instructions and method also include assigning a semantic tag to the parsed input clause, comparing the input clause to a previously received clause, the previously received clause being correlated with a previously generated response clause, and generating an output response message derived from the previously generated response clause.

Claims

1. A method of generating a response to a text-based natural language message, comprising:

identifying a sentence in the text-based natural language message;

identifying an input clause in the sentence;

parsing the input clause, thereby defining a relationship between words in the input clause,

assigning a semantic tag to the parsed input clause;

comparing the input clause to a previously received clause, the previously received clause being correlated with a previously generated response clause; and

generating an output response message derived from the previously generated response clause.

2. A method according to claim 1, further comprising storing the generated output response message, the input clause, and the parsed input clause or the semantic tag in a database.

3. A method according to claim 1, wherein identifying the input clause comprises annotating the sentence with syntactic tags.

4. A method according to claim 1, further comprising identifying a best match input clause to the input clause from a plurality of previously received clauses, the generated output response message being derived from the previously generated response clause that corresponds to the best match input clause.

5. A method according to claim 4, wherein identifying the best match input clause to the input clause further comprises comparing the previously received clause to the input clause on a word-level, the word-level comparison identifying a best match between words from the input clause and words from the previously received clause, or comparing the previously received clause to the input clause on a phrase-level, the phrase-level comparison identifying a best match between a series of contiguous words from the input clause and a series of contiguous words from the previously received clause.

6. A method according to claim 4, further comprising identifying a plurality of clauses associated with the input clause, identifying the best match input clause to the input clause further comprising comparing the previously received clause to the input clause on a tree-level, the tree-level comparison identifying a best match between the plurality of associated clauses and a plurality of clauses associated with the previously received clause.

7. A method according to claim 1, further comprising matching the semantic tag to a historical input tag, the historical input tag associated with the previously generated response clause, the generated output response message being derived from the previously generated response clause associated with the historical input tag.

8. The method according to claim 1, wherein the generated output response message is derived from a plurality of previously generated response clauses, the previously generated response clauses being correlated with the previously received clause.

9. A non-transitory computer-readable storage medium storing instructions that, when executed by a processing device, cause the processing device to generate a response to a text-based natural language message by performing a computer process comprising:

identifying a sentence in the text-based natural language message;

identifying an input clause in the sentence;

parsing the input clause, thereby defining a relationship between words in the input clause,

assigning a semantic tag to the parsed input clause;

comparing the input clause to a previously received clause, the previously received clause being correlated with a previously generated response clause; and

generating an output response message derived from the previously generated response clause.

10. The non-transitory computer-readable storage medium according to claim 9, wherein the computer process further comprises storing the generated output response message, the input clause, and the parsed input clause or the semantic tag in a database.

11. The non-transitory computer-readable storage medium according to claim 9, the computer process further comprising identifying a best match input clause to the input clause from a plurality of previously received clauses, the generated output response message being derived from the previously generated response clause that corresponds to the best match input clause.

12. The non-transitory computer-readable storage medium according to claim 11, wherein the computer process of identifying the best match input clause to the input clause further comprises comparing the previously received clause to the input clause on a word-level, the word-level comparison identifying a best match between words from the input clause and words from the previously received clause, or comparing the previously received clause to the input clause on a phrase-level, the phrase-level comparison identifying a best match between a series of contiguous words from the input clause and a series of contiguous words from the previously received clause.

13. The non-transitory computer-readable storage medium according to claim 11, wherein the computer process further comprises identifying a plurality of clauses associated with the input clause, identifying the best match input clause to the input clause further comprising comparing the previously received clause to the input clause on a tree-level, the tree-level comparison identifying a best match between the plurality of associated clauses and a plurality of clauses associated with the previously received clause.

14. The non-transitory computer-readable storage medium according to claim 11, wherein the computer process further comprises matching the semantic tag to a historical input tag, the historical input tag associated with the previously generated response clause, the generated output response message being derived from the previously generated response clause associated with the historical input tag.

15. A system to generate a response to a text-based natural language message, the system comprising:

a user interface;

a processing device operatively coupled to the user interface; and

a computer-readable storage medium storing instructions that, when executed by a processing device, cause the processing device to generate a response to a text-based natural language message by performing a computer process comprising;

identifying a sentence in the text-based natural language message;

identifying an input clause in the sentence;

parsing the input clause, thereby defining a relationship between words in the input clause,

assigning a semantic tag to the parsed input clause;

comparing the input clause to a previously received clause, the previously received clause being correlated with a previously generated response clause; and

generating an output response message derived from the previously generated response clause.

16. The system defined by claim 15, wherein the computer process further comprises storing the generated output response message, the input clause, and the parsed input clause or the semantic tag in a database.

17. The system defined by claim 15, wherein the computer process further comprises identifying a best match input clause to the input clause from a plurality of previously received clauses, the generated output response message being derived from the previously generated response clause that corresponds to the best match input clause.

18. The system defined by claim 17, wherein the computer process of identifying the best match input clause to the input clause further comprises comparing the previously received clause to the input clause on a word-level, the word-level comparison identifying a best match between words from the input clause and words from the previously received clause, or comparing the previously received clause to the input clause on a phrase-level, the phrase-level comparison identifying a best match between a series of contiguous words from the input clause and a series of contiguous words from the previously received clause.

19. The system defined by claim 17, wherein the computer process further comprises identifying a plurality of clauses associated with the input clause, identifying the best match input clause to the input clause further comprising comparing the previously received clause to the input clause on a tree-level, the tree-level comparison identifying a best match between the plurality of associated clauses and a plurality of clauses associated with the previously received clause.

20. The system defined by claim 15, wherein the computer process further comprises matching the semantic tag to a historical input tag, the historical input tag being associated with the previously generated response clause, the generated output response message being derived from the previously generated response clause associated with the historical input tag.