METHODS TO CREATE A USER PROFILE AND TO SPECIFY A SUGGESTION FOR A NEXT SELECTION OF A USER

• US 20090282034A1
• Filed: 07/22/2009
• Published: 11/12/2009
• Est. Priority Date: 08/30/2002
• Status: Active Grant

• Current Assignee: Sony Deutschland GmbH (Sony Group Corp.)
• Original Assignee: Sony Deutschland GmbH (Sony Group Corp.)
• Inventors: GORONZY, Silke, Kompe, Ralf, Kemp, Thomas, Hying, Christian, Torge, Sunna, Valsan, Zica, Wais, Helmut, Mencl, Robert, Emele, Martin
• Granted Patent: US 7,970,762 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/437 Administration of user profiles, e.g. generation, initialisation, adaptation, distribution
• CPC Class Codes: H04H 60/46 for recognising users' preferences
• CPC Class Codes: H04N 21/25891 being end-user preferences retrieval of video data in a video database based on user preferences G06F16/739; arrangements for recognizing users' preferences H04H60/46; user profiles in network data switching protocols H04L67/306; processing of user preferences or user profiles in wireless networks H04W8/18
• CPC Class Codes: H04N 21/44222 Analytics of user selections, e.g. selection of programs or purchase activity monitoring of user selections in data processing systems G06F11/34; arrangements for monitoring the user's behaviour or opinions in broadcast systems H04H60/33
• CPC Class Codes: H04N 21/6582 Data stored in the client, e.g. viewing habits, hardware capabilities, credit card number arrangements where receivers interact with the broadcast H04H20/38
• CPC Class Codes: H04N 21/8106 involving special audio data, e.g. different tracks for different languages
• CPC Class Codes: H04N 7/163 by receiver means only
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation

Abstract

A user profile and/or the suggestions computed based thereon are obtained taking a special set of user features into account. The user features are defined to represent a typical general behaviour of an individual user in respect to the application where the user profile is used. In other words, for each application where a user profile is used a special set of user features are defined which are able to represent a typical general behaviour of an individual user. Based on these user features the weights in the list of word-weight pairs or weighted keywords which represents the user profile are computed or influenced during the creation of the user profile, and/or a mufti-user profile is split during the creation of an individual user profile from a mufti-user profile, and/or during specification of a suggestion a user history which is used to create the user profile, and/or the user profile, and/or the suggestion results are filtered.

Claims

1. A method for providing program suggestions to a user based on electronic program guides (EPGs), comprising:

retrieving electronic program guide articles from an electronic program guide articles database; and

suggesting at least one electronic program guide article to the user based on the retrieved electronic program guide articles and based on a user profile,

wherein the user profile corresponds to a generic stereotype user profile, when providing a suggestion for the first time, to a linear combination of the generic stereotype user profile, at least one specific stereotype profile, and an individual user profile during a first predetermined period during the collection of data, and to the individual user profile afterwards, wherein during the first predetermined period, a first weighting factor of the at least one specific stereotype profile is first increased and then decreased and a second weighting factor of the generic stereotype user profile is continuously decreased,

wherein the generic stereotype user profile describes an average interest in everything and the specific stereotype user profile defines interests of a user who focuses on a specific topic, and

wherein the user profile comprises a list of word-weight pairs, wherein the weights are computed based on user features that represent a typical general behaviour.

2. The method according to claim 1, wherein the user features are influenced from a user history.

3. The method according to claim 2, wherein the user history comprises electronic program guide articles corresponding to programs that the user has previously watched.

4. The method according to claim 2, wherein the user features comprise features related to the previous typical user behaviour.

5. The method according to claim 1, wherein the user features comprise one or more of the following features:

preferred channel of audio/video program consumed by the user;

typical time to consume an audio/video program by the user;

length of consuming an audio/video program by the user in relation to the total length of the audio/video program;

time of beginning the consuming an audio/video program by the user in relation to a start time of the audio/video program;

typical length of consuming an audio video program by the user in relation to the time of consuming;

relation between how often a particular audio/video program is consumable and how often it is consumed by the user;

general audio/video program consuming behaviour of the user, in particular in relation to a switch-on time and length of a used audio/video device;

audio/video programs recorded by the user;

time duration between the recording of a particular audio/video program by the user and the consuming of the audio/video program by the user;

actual mood of the user;

actual wish of audio/video program entered by the user;

year of production of an audio/video program consumed by the user;

director and/or actor and/or group of actors of an audio/video program consumed by the user;

type of an audio/video program consumed by the user; and

/or

title of an audio/video program consumed by the user.

6. The method according to claim 1, further comprising:

dividing the user profile into sub user profiles according to characteristics of stereotype user profiles.

7. The method according to claim 1, further comprising:

determining words to be included in the list of word-weight pairs on basis of words included in program descriptions of previous selections by the user.

8. The method according to claim 7, further comprising:

determining further words to be included in the list of word-weight pairs on basis of a database evaluation of co-occurrences of words already included in the list.

9. The method according to claim 7, further comprising:

computing a weight of a word to be included in the list of word-weight pairs on a basis of a comparison of an inverse document frequency of the word with respect to all available program descriptions and an inverse document frequency of the word with respect to program descriptions of previous selections by the user.

10. The method according to claim 7, further comprising:

computing weights of words to be included in the list of word-weight pairs as a product of;

importance of the word with respect to all program descriptions of previous selections by the user,

and at least one of the following terms;

a constant term, in particular one;

importance of the word with respect to the program description in which the word is included;

importance of the word with respect to all available program descriptions; and

a correctional factor that depends on the length of the program description in which the word is included and the frequency of the word in this program description.

11. The method according to claim 10, wherein the importance of a word with respect to all program descriptions of previous selections by the user is determined on a basis of the portion of the word in the set of all words of the program descriptions of previous selections by the user.

12. The method according to claim 11, wherein the importance of a word with respect to all program descriptions of previous selections by the user is determined under consideration of the program description in which the word is included.

13. The method according to claim 10, wherein the importance of the word with respect to the program description in which the word is included is determined on a basis of occurrences of the word in the program description.

14. The method according to claim 10, wherein the importance of the word with respect to all available program descriptions is determined on basis of its inverse document frequency in respect to all available program descriptions.

15. The method according to claim 14, wherein the importance of a word with respect to all available program descriptions is determined under consideration of the program description in which the word is included.

16. The method according to claim 10, further comprising:

computing weights of words to be included in the list of word-weight pairs according to the following formula;

17. The method according to claim 10, further comprising:

normalizing a weight of a word included in the list of word-weight pairs by a behaviour of the user to stay with a selection and to switch selections off or to switch on/off just for this selection.

18. The method according to claim 10, further comprising:

normalizing a weight of a word included in the list of word-weight pairs by a behaviour of the user when to actually consume a recorded selection.

19. The method according to claim 10, further comprising:

averaging the weights w_i(t) of a word over all program descriptions of previous selections by the user in which the word occurs to a weight w(t) of the new profile.

20. The method according to claim 10, further comprising:

excluding words from the list of word-weight pairs which show a weight below a predetermined threshold or within a predetermined lower range of weights.

21. The method according to claim 9, further comprising:

computing a weight of a word to be included in the list of word-weight pairs by considering negative selections of the user.

22. The method according to claim 9, further comprising:

computing a weight of a word to be included in the list of word-weight pairs by considering a maximum frequency of possible user selections in respect to certain user features.

23. The method according to claim 7, wherein the created user profile is used as a query in possible future program descriptions to suggest at least one possible future selection to the user.

24. The method according to claim 17, wherein for each possible future program description a modified OKAPI weight is computed, wherein each matching word has its weight as a co-factor so that the search result is influenced according to the user profile.

25. Method according to claim 24, wherein the modified OKAPI weight is computed according to the following formula:

26. The method according to claim 24, wherein the modified OKAPI weight is computed according to the following formula;

27. The method according to claim 1, wherein a weight with which a specific stereotype user profile is considered during the linear combination to provide the suggestion is computed by matching the user selections with the data in the respective specific stereotype and increasing the weight of the stereotype if a match is found.

28. The method according to claim 1, wherein

the user profile (2) is created from a multi-user profile that comprises a list of word-weight pairs,

wherein the multi-user profile is splitted at least once according to the user features.

29. The method according to claim 28, wherein for splitting the multi-user profile the following are performed:

performing a tentative split according to user features to generate a first and a second sub user profiles;

calculating the relative difference between the two sub user profiles;

performing a tentative split and calculating the relative difference until all or a predetermined number of tentative splits are performed; and

splitting the multi-user profile according to a tentative split that yields a highest relative difference in case the relative difference lies above a predetermined threshold.

30. The method according to claim 29, wherein the relative difference is calculated by calculating a difference of a first discrete probability distribution of the first sub user profile over the user features that are contained therein and of a second discrete probability distribution of the second sub user profile over the user features that are contained therein.

31. The method according to claim 30, wherein the difference of the two discrete probability distributions is calculated using the symmetrized Kulback Leibler-distance sum, where events which happen zero times are replaced by one virtual occurrence or where only events which happen at least once in both distributions are taken into account.

32. The method according to claim 1, wherein the general key structure includes a forgetting factor.

33. The method according to claim 1, wherein a future program comprises a stored personal content.

34. The method according to claim 1, wherein it is used in an audio/video program suggestion engine.

35. The method according to claim 34, wherein the audio/video program suggestion engine is internet based.

36. The method according to claim 1, wherein it is client based.

37. A computer program product, comprising computer program means adapted to perform the method as defined in claim 1 when being executed on a computer, micro processor, digital signal processor, including home server, set-top-box, TV, VCR, PDA.

38. A computer readable storage medium, storing thereon a computer program product according to claim 37.

39. A suggestion engine to specify a suggestion for a next selection of a user, which suggestion is determined based on a method as defined in claim 1.