Inferring User Intentions Based on User Conversation Data and Spatio-Temporal Data

• US 20170249956A1
• Filed: 02/29/2016
• Published: 08/31/2017
• Est. Priority Date: 02/29/2016
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Dey, Prasenjit, Ekambaram, Vijay, Joshi, Vikas V., Narayanam, Ramasuri
• Granted Patent: US 9,905,248 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 16/337 Profile generation, learning or modification
• CPC Class Codes: G06F 40/295 Named entity recognition
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/24 Speech recognition using non-acoustical features
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 2015/228 of application context
• CPC Class Codes: G10L 25/48 specially adapted for particular use

Abstract

Systems and methods are provided for inferring explicit user intentions based on user conversation data and associated spatial-temporal data of the user. For example, a method includes detecting a named entity in a conversation of a user communicating on a mobile computing device of the user, obtaining spatio-temporal information associated with the user, wherein the spatio-temporal information comprises information regarding one or more locations of the user and timing of the user'"'"'s presence at the one or more locations, and utilizing the spatio-temporal information to disambiguate the named entity detected in the conversation of the user. The disambiguated named entity and the spatio-temporal data associated with the user are utilized to determine user intentions and to generate or update user profile information associated with the user.

Claims

1-17. -17. (canceled)

18. A method, comprising steps of:

temporarily buffering one or more portions of a voice conversation of a user communicating on a mobile computing device of the user;

performing automatic speech recognition on the buffered portions of the voice conversation to detect an entity named during the voice conversation of the user;

obtaining spatio-temporal information corresponding to the voice conversation of the user, the spatio-temporal information comprising information regarding (i) one or more locations of the user and (ii) timing of the user'"'"'s presence at the one or more locations, wherein obtaining the spatio-temporal information comprises;

determining spatio-temporal contextual information from at least one of (i) one or more keywords and (ii) one or more phrases in the voice conversation of the user which are detected as a result of the automatic speech recognition on the buffered portions of the voice conversation; and

obtaining spatio-temporal information from one or more location services executing on the mobile computing device based on the determined spatio-temporal contextual information; and

utilizing the spatio-temporal information to disambiguate the named entity detected in the conversation of the user;

wherein the steps are performed by executing program code on the mobile computing device.

19. The method of claim 18, wherein said determining spatio-temporal contextual information and said obtaining spatio-temporal information based on the determined spatio-temporal contextual information comprises:

determining a timeframe based on temporal contextual information determined from the voice conversation of the user; and

obtaining global positioning system data corresponding to the determined timeframe;

wherein the global positioning system data are utilized to disambiguate the named entity detected in the conversation of the user.

20. The method of claim 19, wherein said obtaining global positioning system data corresponding to the determined timeframe comprises at least one of (i) obtaining previously collected global positioning system data corresponding to the determined timeframe and (ii) setting a trigger to collect global positioning system data at a future time corresponding to the determined timeframe.

21. The method of claim 18, comprising:

determining a user intention based on (i) the disambiguated named entity and (ii) the spatio-temporal information associated with the user.

22. The method of claim 21, wherein the user intention comprises a reason for the user visiting a given location.

23. The method of claim 18, wherein said detecting a named entity during the conversation of the user comprises detecting at least one of (i) one or more keywords and (ii) one or more phrases in a given block of text data obtained from transcribing a voice communication of the user into text via the automatic speech recognition.

24. An article of manufacture comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor of a mobile computing device to cause the mobile computing device to perform a method comprising:

temporarily buffering one or more portions of a voice conversation of a user communicating on a mobile computing device of the user;

performing automatic speech recognition on the buffered portions of the voice conversation to detect an entity named during the voice conversation of the user;

obtaining spatio-temporal information corresponding to the voice conversation of the user, the spatio-temporal information comprising information regarding (i) one or more locations of the user and (ii) timing of the user'"'"'s presence at the one or more locations, wherein obtaining the spatio-temporal information comprises;

determining spatio-temporal contextual information from at least one of (i) one or more keywords and (ii) one or more phrases in the voice conversation of the user which are detected as a result of the automatic speech recognition on the buffered portions of the voice conversation; and

obtaining spatio-temporal information from one or more location services executing on the mobile computing device based on the determined spatio-temporal contextual information; and

utilizing the spatio-temporal information to disambiguate the named entity detected in the conversation of the user.

25. The article of manufacture of claim 24, wherein said determining spatio-temporal contextual information and said obtaining spatio-temporal information based on the determined spatio-temporal contextual information comprises:

determining a timeframe based on temporal contextual information determined from the voice conversation of the user; and

obtaining global positioning system data corresponding to the determined timeframe;

wherein the global positioning system data are utilized to disambiguate the named entity detected in the conversation of the user.

26. The article of manufacture of claim 25, wherein said obtaining global positioning system data corresponding to the determined timeframe comprises at least one of (i) obtaining previously collected global positioning system data corresponding to the determined timeframe and (ii) setting a trigger to collect global positioning system data at a future time corresponding to the determined timeframe.

27. The article of manufacture of claim 24, wherein the program instructions further cause the mobile computing device to:

determine a user intention based on (i) the disambiguated named entity and (ii) the spatio-temporal information associated with the user.

28. The article of manufacture of claim 27, wherein the user intention comprises a reason for the user visiting a given location.

29. The article of manufacture of claim 24, wherein said detecting a named entity during the conversation of the user comprises detecting at least one of (i) one or more keywords and (ii) one or more phrases in a given block of text data obtained from transcribing a voice communication of the user into text via the automatic speech recognition.

30. A mobile computing device, comprising:

a memory configured to store program instructions; and

a processor configured to execute the program instructions stored in the memory to implement a method which comprises;

temporarily buffering one or more portions of a voice conversation of a user communicating on a mobile computing device of the user;

performing automatic speech recognition on the buffered portions of the voice conversation to detect an entity named during the voice conversation of the user;

obtaining spatio-temporal information corresponding to the voice conversation of the user, the spatio-temporal information comprising information regarding (i) one or more locations of the user and (ii) timing of the user'"'"'s presence at the one or more locations, wherein obtaining the spatio-temporal information comprises;

determining spatio-temporal contextual information from at least one of (i) one or more keywords and (ii) one or more phrases in the voice conversation of the user which are detected as a result of the automatic speech recognition on the buffered portions of the voice conversation; and

obtaining spatio-temporal information from one or more location services executing on the mobile computing device based on the determined spatio-temporal contextual information; and

utilizing the spatio-temporal information to disambiguate the named entity detected in the conversation of the user.

31. The mobile computing device of claim 30, wherein said determining spatio-temporal contextual information and said obtaining spatio-temporal information based on the determined spatio-temporal contextual information comprises:

determining a timeframe based on temporal contextual information determined from the voice conversation of the user; and

obtaining global positioning system data corresponding to the determined timeframe;

wherein the global positioning system data are utilized to disambiguate the named entity detected in the conversation of the user.

32. The mobile computing device of claim 31, wherein said obtaining global positioning system data corresponding to the determined timeframe comprises at least one of (i) obtaining previously collected global positioning system data corresponding to the determined timeframe and (ii) setting a trigger to collect global positioning system data at a future time corresponding to the determined timeframe.

33. The mobile computing device of claim 30, wherein the processor configured to execute the program instructions stored in the memory to implement a method which further comprises:

determining a user intention based on (i) the disambiguated named entity and (ii) the spatio-temporal information associated with the user.

34. The mobile computing device of claim 33, wherein the user intention comprises a reason for the user visiting a given location.

35. The mobile computing device of claim 30, wherein said detecting a named entity during the conversation of the user comprises detecting at least one of (i) one or more keywords and (ii) one or more phrases in a given block of text data obtained from transcribing a voice communication of the user into text via the automatic speech recognition.