Natural language processing for a location-based services system

• US 7,437,295 B2
• Filed: 04/25/2002
• Issued: 10/14/2008
• Est. Priority Date: 04/27/2001
• Status: Expired due to Term

• Current Assignee: Accenture Global Services Limited (Accenture PLC)
• Original Assignee: Accenture LLP (Accenture PLC)
• Inventors: Che, Vinny Wai-yan, Dempsen, Stephen L., Pitts, Ashton F. III
• Primary Examiner(s): Dorvil; Richemond
• Assistant Examiner(s): YEN, ERIC L
• Application Number:
  US10/131,898
• Publication Number:
  US 20020161587A1
• Time in Patent Office: 2,364 Days
• Field of Search: 704/270.1, 704/276, 704/9, 704/270, 704/275, 379/58, 364/900, 455/422.1
• US Class Current: 704/275
• CPC Class Codes: G06F 16/24522 Translation of natural language queries to structured queries
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 16/9537 Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G10L 15/1822 Parsing for meaning understanding
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: G10L 2015/228 of application context
• CPC Class Codes: H04L 67/306 User profiles
• CPC Class Codes: H04L 67/52 specially adapted for the location of the user terminal
• CPC Class Codes: H04M 2201/40 using speech recognition
• CPC Class Codes: H04M 2207/18 wireless networks
• CPC Class Codes: H04M 2242/15 Information service where the information is dependent on the location of the subscriber
• CPC Class Codes: H04M 2242/30 Determination of the location of a subscriber
• CPC Class Codes: H04M 3/42204 Arrangements at the exchange for service or number selection by voice at the terminal H04M1/27
• CPC Class Codes: H04M 3/42348 Location-based services which utilize the location information of a target
• CPC Class Codes: H04M 3/4936 Speech interaction details speech recognition per se G10L15/00
• CPC Class Codes: H04W 4/02 Services making use of location information
• CPC Class Codes: H04W 4/029 Location-based management or tracking services
• CPC Class Codes: H04W 4/20 : Services signaling; Auxiliary data signalling, i.e. transmitting data via a non-traffic channel
• CPC Class Codes: H04W 4/23 : for mobile advertising

Abstract

A method and system for providing natural language processing in a communication system is disclosed. A voice request is generated with a remote terminal that is transmitted to a base station. A speech recognition application is then used to identify a plurality of words that are contained in the voice request. After the words are identified, a grammar associated with each word is also identified. Once the grammars have been identified, each word is categorized into a respective grammar category. A structured response is then generated to the voice request with a response generation application.

Claims

1. A method of providing natural language processing in a communication system, comprising the steps of:

receiving a non-predefined voice request for business information related to a product, and a geographic location;

identifying a plurality of words contained in said non-predefined voice request for information with a speech recognition application;

identifying a grammar associated with each said word, wherein said grammar is selected from a group of grammars consisting of a noun, an adjective, an adverb, a preposition, a conjunction, a pronoun, a verb, and an interjection;

categorizing each said word into a respective grammar category;

generating a structured request based on said geographic location and said categorization of said words which comprises ignoring each said word that is identified as a pronoun, limiting the structured request based on a meaning of at least one of said words that is identified as a noun, and restricting the structured request for each said word that is identified as a conjunction;

searching a grammar database for a similar request that contains words in the same grammar category as said structured request;

when said similar request is not found in said grammar database, searching a plurality of business data files for information to be used in a structured response, wherein said business data files include product and service related information of a plurality of different businesses;

updating said grammar database to include said structured request and said structured response to avoid a future search of said business data files for a similar structured request; and

generating said structured response to said non-predefined voice request for information with a response generation application.

2. The method of claim 1, further comprising the step of narrowing said structured response to a geographic area in which a remote terminal is located based on said received geographic location.

3. The method of claim 1, wherein said voice request contains a request for a particular product.

4. The method of claim 1, wherein said structured response is generated as a voice response.

5. The method of claim 1, wherein said structured response is generated as a text message.

6. The method of claim 1, further comprises the step of transmitting said structured response to a remote terminal.

7. The method of claim 1, wherein categorizing each said word into a respective grammar category comprises categorizing each said word into one of a plurality of grammar categories based on a grammatical use of each said word in said non-predefined voice request, the grammar categories being noun, adjective, adverb, preposition, conjunction, pronoun, verb, and interjection.

8. The method of claim 1, wherein searching comprises searching based on said structured request and said geographic location.

9. The method of claim 1, wherein receiving said non-predefined voice request for business information related to a product comprises receiving said geographic location at about the same time as said non-predefined voice request for business information is received.

10. The method of claim 1, wherein receiving said non-predefined voice request for business information related to a product comprises receiving said geographic location prior to receipt of said non-predefined voice request for business information.

11. A wireless communication system with natural language processing, comprising:

a speech recognition application configured to receive a non-predefined voice request for information from a remote terminal, wherein said speech recognition application is operable to identify a plurality of words contained in said non-predefined voice request for information;

a natural language processing application connected to said speech recognition application, wherein said natural language processing application is operable to generate a structured request based on said words contained in said non-predefined voice request for information and a geographic location of said remote terminal;

wherein said natural language processing application is operable to categorize said words contained in said non-predefined voice request for information into a respective one of a plurality of grammar categories;

wherein said words in said grammar categories and said geographic location are useable by a response generation application connected to said natural language processing application to locate in a grammar database a product identified in said non-predefined voice request for information;

wherein said response generation application is further operable to retrieve said product identified in said non-predefined voice request for information, based on said geographic location, from a plurality of business data files when said product is not locatable in said grammar database; and

wherein said response generation application is further operable to update said grammar database with said retrieved product, and generate a structured response to said structured request.

12. The wireless communication system of claim 11, further comprising a voice synthesis application connected to said response generation application for transforming said structured response into a voice response.

13. The wireless communication system of claim 11, wherein said non-predefined voice request for information contains a request for a particular product.

14. The wireless communication system of claim 11, wherein said structured response is generated as a text message that is sent to said remote terminal.

15. The wireless communication system of claim 11, wherein said geographic location is generatable with said remote terminal.

16. The wireless communication system of claim 15, wherein said geographic location is pre-settable with said remote terminal to a desired geographic location.

17. The wireless communication system of claim 11, wherein said geographic location is generatable with a base station in communication with said remote terminal.

18. The wireless communication system of claim 11, wherein said response generation application is operable to search said grammar database for the same words in the same grammar categories to locate said product in said grammar database.

24. The method of claim 17 further comprises the step of transmitting said structured response to said remote terminal.

19. A method of providing natural language processing in a location-based services system, comprising the steps of:

receiving a non-predefined voice request for information from a remote terminal, and a geographic location of said remote terminal;

identifying a plurality of words contained in said non-predefined voice request for information with a speech recognition application;

identifying a grammar associated with each said word;

categorizing each said word into a respective grammar category based on grammatical use;

generating a structured request based on said categorization of said words;

searching for matches to each said categorized word in said structured request in a grammar database with a response generation application, said grammar database organized based on parts of speech into a plurality of grammar categories each containing a plurality of words that are searchable for matches to each said categorized word that are in said respective grammar category;

searching for matches to each said categorized word in a database of products with said response generation application only when there are no matches found in said grammar database, said database of products containing business data files businesses listed within said location-based services system;

generating a structured response to said non-predefined voice request for information with said response generation application, wherein said structured response is geographically targeted based on said geographic location of said remote terminal; and

in response to matches to each said categorized word in said database of products, updating said grammar database to include said structure request and said structured response to avoid a future search in said database of products for a similar structured request.

20. The method of claim 19, wherein said non-predefined voice request for information contains a request for a particular product.

21. The method of claim 19, wherein the grammar is selected from a group of grammars consisting of a noun, an adjective, an adverb, a preposition, a conjunction, a pronoun, a verb and an interjection.

22. The method of claim 19, wherein said structured response is generated as a voice response.

23. The method of claim 19, wherein said structured response is generated as a text message.

25. The method of claim 19, wherein categorizing each said word into a respective grammar category comprises categorizing each said word into one of a plurality of grammar categories based on a grammatical use of each said word in said non-predefined voice request, the grammmar categories being noun, adjective, adverb, preposition, conjunction, pronoun, verb, and interjection.

26. The method of claim 19, wherein receiving said non-predefined voice request comprises receiving said geographic location at about the same time as said non-predefined voice request for information.

27. The method of claim 19, wherein receiving said non-predefined voice request comprises receiving said geographic location prior to receipt of said non-predefined voice request.