System and method for selecting and presenting advertisements based on natural language processing of voice-based input

• US 8,145,489 B2
• Filed: 07/30/2010
• Issued: 03/27/2012
• Est. Priority Date: 02/06/2007
• Status: Active Grant

• Current Assignee: VB Assets, LLC
• Original Assignee: VoiceBox Technologies, Inc. (Microsoft Corporation)
• Inventors: Kennewick, Mike, Freeman, Tom
• Primary Examiner(s): ALBERTALLI, BRIAN LOUIS
• Application Number:
  US12/847,564
• Publication Number:
  US 20100299142A1
• Time in Patent Office: 606 Days
• Field of Search: None
• US Class Current: 704/257
• CPC Class Codes: G06F 40/263 Language identification
• CPC Class Codes: G06Q 30/0241 Advertisements
• CPC Class Codes: G06Q 30/0242 Determining effectiveness of advertisements
• CPC Class Codes: G06Q 30/0251 Targeted advertisements
• CPC Class Codes: G06Q 30/0269 based on user profile or attribute
• CPC Class Codes: G10L 15/18 using natural language modelling
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

A system and method for selecting and presenting advertisements based on natural language processing of voice-based inputs is provided. A user utterance may be received at an input device, and a conversational, natural language processor may identify a request from the utterance. At least one advertisement may be selected and presented to the user based on the identified request. The advertisement may be presented as a natural language response, thereby creating a conversational feel to the presentation of advertisements. The request and the user'"'"'s subsequent interaction with the advertisement may be tracked to build user statistical profiles, thus enhancing subsequent selection and presentation of advertisements.

Claims

1. A method for selecting and presenting advertisements in response to processing natural language utterances, comprising:

recognizing, at a server, one or more words or phrases in a natural language utterance that contains a request, wherein the server includes a speech recognition engine that recognizes the one or more words or phrases in the natural language utterance;

interpreting the recognized words or phrases at a conversational language processor coupled to the server and the speech recognition engine to establish a context associated with the natural language utterance;

delivering an advertisement in the established context associated with the natural language utterance from the server to an electronic device in communication with the server, wherein the advertisement relates to one or more of the recognized words or phrases, an action associated with the request, a personalized cognitive model derived from an interaction pattern associated with a specific user, a generalized cognitive model derived from an interaction pattern associated with a plurality of users, or an environmental model derived from environmental conditions or surroundings associated with the specific user;

tracking, at the server, an interaction pattern associated with the advertisement delivered to the electronic device;

updating, at the server, the personalized cognitive model, the generalized cognitive model, and the environmental model based on the tracked interaction pattern associated with the advertisement;

building, at the server, long-term shared knowledge and short-term shared knowledge from the updated personalized cognitive model, the updated generalized cognitive model, and the updated environmental model; and

interpreting a subsequent natural language utterance at the conversational language processor using the long-term shared knowledge and the short-term shared knowledge.

2. The method of claim 1, wherein the updated personalized cognitive model, the updated generalized cognitive model, and the updated environmental model provide statistical profiles to select one or more subsequent advertisements in response to the subsequent natural language utterance.

3. The method of claim 2, wherein the statistical profiles identify one or more affinities between the advertisement delivered to the electronic device and one or more of the recognized words or phrases, the action associated with the request, the updated personalized cognitive model, the updated generalized cognitive model, or the updated environmental model.

4. The method of claim 1, wherein the tracked interaction pattern associated with the advertisement includes information associated with a subsequent request in which the electronic device interacts with the advertisement.

5. The method of claim 4, wherein the subsequent request in which the electronic device interacts with the advertisement includes the electronic device initiating a task or initiating a query relating to the advertisement.

6. The method of claim 1 further comprising using, at the conversational language processor, the tracked interaction pattern associated with the advertisement to resolve the request in response to determining that the natural language utterance includes incomplete or ambiguous information.

7. The method of claim 1, wherein the speech recognition engine recognizes the one or more words or phrases in the natural language utterance using a plurality of dictionary and phrase entries that are dynamically updated based on a history of a current dialog and one or more prior dialogs.

8. The method of claim 7, wherein the plurality of dictionary and phrase entries are further dynamically updated based on one or more dynamic fuzzy set probabilities or prior probabilities derived from the history of the current dialog and the prior dialogs.

9. A system to select and present advertisements in response to natural language utterances, comprising:

a speech recognition engine configured to recognize one or more words or phrases in a natural language utterance that contains a request;

a conversational language processor configured to interpret the recognized words or phrases to establish a context associated with the natural language utterance; and

a server associated with the speech recognition engine and the conversational language processor, wherein the server is configured to;

deliver an advertisement in the established context associated with the natural language utterance to an electronic device in communication with the server, wherein the advertisement relates to one or more of the recognized words or phrases, an action associated with the request contained in the utterance, a personalized cognitive model derived from an interaction pattern associated with a specific user, a generalized cognitive model derived from an interaction pattern associated with a plurality of users, or an environmental model derived from environmental conditions or surroundings associated with the specific user;

track an interaction pattern associated with the advertisement delivered to the electronic device;

update the personalized cognitive model, the generalized cognitive model, and the environmental model based on the tracked interaction pattern associated with the advertisement; and

build long-term shared knowledge and, short-term shared knowledge from the updated personalized cognitive model, the updated generalized cognitive model, and the updated environmental model, wherein the conversational language processor is further configured to use the long-term shared knowledge and the short-term shared knowledge to interpret a subsequent natural language utterance.

10. The system of claim 9, wherein the updated personalized cognitive model, the updated generalized cognitive model, and the updated environmental model provide statistical profiles to select one or more subsequent advertisements in response to the subsequent natural language utterance.

11. The system of claim 10, wherein the statistical profiles identify one or more affinities between the advertisement delivered to the electronic device and one or more of the recognized words or phrases, the action associated with the request, the updated personalized cognitive model, the updated generalized cognitive model, or the updated environmental model.

12. The system of claim 9, wherein the tracked interaction pattern associated with the advertisement includes information associated with a subsequent request in which the electronic device interacts with the advertisement.

13. The system of claim 12, wherein the subsequent request in which the electronic device interacts with the advertisement includes a task that the electronic device initiated or a query that the electronic device initiated in relation to the advertisement.

14. The system of claim 9 wherein the conversational language processor is further configured to use the tracked interaction pattern associated with the advertisement to resolve the request in response to the natural language utterance including incomplete or ambiguous information.

15. The system of claim 9 wherein the speech recognition engine is configured to use a plurality of dictionary and phrase entries that are dynamically updated based on a history of a current dialog and one or more prior dialogs to recognize the one or more words or phrases in the natural language utterance.

16. The system of claim 15 wherein the plurality of dictionary and phrase entries are further dynamically updated based on one or more dynamic fuzzy set probabilities or prior probabilities derived from the history of the current dialog and the prior dialogs.

17. A method for selecting and presenting advertisements in response to processing natural language utterances, comprising:

recognizing, at a speech recognition engine that executes on a server, one or more words or phrases in a natural language utterance;

interpreting the one or more recognized words or phrases at a conversational language processor that executes on the server to establish a context associated with the natural language utterance;

delivering an advertisement in the established context associated with the natural language utterance from the server to an electronic device in communication with the server;

tracking, at the server, an interaction pattern associated with the advertisement delivered to the electronic device;

updating, at the server, one or more models based on the tracked interaction pattern associated with the advertisement, wherein the one or more updated models include updated information associated with a specific user, updated information associated with a plurality of users, and updated information relating to environmental conditions or surroundings associated with the specific user; and

interpreting a subsequent natural language utterance at the conversational language processor using long-term shared knowledge and short-term shared knowledge derived from the updated information associated with the one or more updated models.

18. A system to select and present advertisements in response to natural language utterances, comprising:

a speech recognition engine configured to recognize one or more words or phrases in a natural language utterance;

a conversational language processor configured to interpret the one or more recognized words or phrases to establish a context associated with the natural language utterance; and

a server associated with the speech recognition engine and the conversational language processor, wherein the server is configured to;

deliver an advertisement in the established context associated with the natural language utterance to an electronic device in communication with the server;

track an interaction pattern associated with the advertisement delivered to the electronic device; and

update, based on the tracked interaction pattern associated with the advertisement, one or more models to include updated information associated with a specific user, updated information associated with a plurality of users, and updated information relating to environmental conditions or surroundings associated with the specific user, wherein the conversational language processor is further configured to use long-term shared knowledge and short-term shared knowledge derived from the updated information associated with the one or more updated models to interpret a subsequent natural language utterance.