System, method and computer program product for predicting item preference using revenue-weighted collaborative filter

• US 10,210,534 B2
• Filed: 10/19/2016
• Issued: 02/19/2019
• Est. Priority Date: 06/30/2011
• Status: Active Grant

• Current Assignee: TrueCar, Inc.
• Original Assignee: TrueCar, Inc.
• Inventors: Sullivan, Thomas J., Swinson, Michael D.
• Primary Examiner(s): St Leger, Geoffrey R
• Assistant Examiner(s): Luu, Cuong V
• Application Number:
  US15/297,883
• Publication Number:
  US 20170039580A1
• Time in Patent Office: 853 Days
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 3/0482 Interaction with lists of selectable items, e.g. menus
• CPC Class Codes: G06F 3/04842 Selection of displayed objects or displayed text elements G06F3/0482 takes precedence
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities

Abstract

Embodiments disclosed provide a system, method, and computer program product for identifying consumer items more likely to be bought by an individual user. In some embodiments, a collaborative filter may be used to rank items based on the degree to which they match user preferences. The collaborative filter may be hierarchical and may take various factors into consideration. Example factors may include the similarity among items based on observable features, a summary of aggregate online search behavior across multiple users, the item features determined to be most important to the individual user, and a baseline item against which a conditional probability of another item being selected is measured.

Claims

1. A method for identifying consumer items more likely to be bought by a user, comprising:

in a computer having a processor and a memory, wherein the computer is communicatively connected to a computing device;

generating an interface which is presented to the user via the computing device;

determining, responsive to the user viewing a baseline item via the interface on the computing device, for each individual item of a plurality of items, a probability that the user will select the individual item, given that the user has expressed interest in the baseline item, wherein the probability is determined based on;

a similarity among the plurality of items based on observable features thereof;

an aggregate online search behavior relative to individual items across multiple users; and

preferences of the user with respect to the baseline item, wherein the baseline item is viewed by the user; and

ranking, in decreasing order, probabilities determined for the plurality of items, wherein items that are more likely to be bought by the user are ranked higher than items that are less likely to be bought by the user;

determining a probability that the user will select a particular one of the individual items of the plurality of items after the particular baseline item is viewed by the user; and

updating the interface running on the computing device, wherein the updated interface presents to the user at least a portion of the plurality of items ranked in decreasing order according to the probabilities.

2. The method according to claim 1, further comprising determining the similarity among the plurality of items, including:

a) computing individual feature difference between a first observation and a second observation;

b) computing a composite similarity between the first observation and the second observation; and

c) repeating a) and b) for all possible values of the first observation and the second observation.

3. The method according to claim 1, further comprising collecting item view frequencies only for each hop by the user in a sequence of item discovery.

4. The method according to claim 1, further comprising collecting item view frequencies only for all hops by the user in a sequence of item discovery.

5. The method according to claim 1, further comprising collecting item view frequencies only for all pairs of items in a sequence of item discovery.

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

for each paired observation associated with a baseline item, determining what a first next item is expected to be selected by the user; and

examining all observations associated with the baseline item to determine what a second next item is expected to be selected by the user.

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

defining a kernel for a given radius each time a second next item is selected after a baseline item is established by the user.

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

making a prediction as to which next item the user will select;

comparing the prediction with an actual next item selected by the user; and

assigning a penalty value to the prediction if the prediction is incorrect.

9. The method according to claim 8, wherein the prediction is one of a plurality of predictions, further comprising:

determining a total penalty value for all incorrect predictions in the plurality of predictions; and

determining a set of weights that minimizes the total penalty.

10. A computer program product comprising at least one non-transitory computer readable medium storing instructions translatable by at least one processor in a computer having a memory, wherein the computer is communicatively connected to a computing device, wherein the instructions are translatable by the at least one processor to perform:

generating an interface which is presented to a user via the computing device;

determining, responsive to the user viewing a baseline item via the interface on the computing device, for each individual item of a plurality of items, a probability that the user will select the individual item, given that the user has expressed interest in the baseline item, wherein the probability is determined based on;

a similarity among the plurality of items based on observable features thereof;

an aggregate online search behavior relative to individual items across multiple users; and

preferences of the user with respect to the baseline item, wherein the baseline item is viewed by the user;

ranking, in decreasing order, probabilities determined for the plurality of items, wherein items that are more likely to be bought by the user are ranked higher than items that are less likely to be bought by the user;

determining a probability that the user will select a particular one of the individual items of the plurality of items after the particular baseline item is viewed by the user; and

updating the interface running on the computing device, wherein the updated interface presents to the user at least a portion of the plurality of items ranked in decreasing order according to the probabilities.

11. The computer program product of claim 10, wherein the at least one non-transitory computer readable medium further stores instructions translatable by the at least one processor to perform:

determining the similarity among the plurality of items, including;

a) computing individual feature difference between a first observation and a second observation;

b) computing a composite similarity between the first observation and the second observation; and

c) repeating a) and b) for all possible values of the first observation and the second observation.

12. The computer program product of claim 10, wherein the at least one non-transitory computer readable medium further stores instructions translatable by the at least one processor to perform:

collecting item view frequencies for each hop by the user in a sequence of item discovery.

13. The computer program product of claim 10, wherein the at least one non-transitory computer readable medium further stores instructions translatable by the at least one processor to perform:

collecting item view frequencies for all hops by the user in a sequence of item discovery.

14. The computer program product of claim 10, wherein the at least one non-transitory computer readable medium further stores instructions translatable by the at least one processor to perform:

collecting item view frequencies for all pairs of items in a sequence of item discovery.

15. The computer program product of claim 10, wherein the at least one non-transitory computer readable medium further stores instructions translatable by the at least one processor to perform:

for each paired observation associated with a baseline item, determining what a first next item is expected to be selected by the user; and

examining all observations associated with the baseline item to determine what a second next item is expected to be selected by the user.

16. The computer program product of claim 15, wherein the at least one non-transitory computer readable medium further stores instructions translatable by the at least one processor to perform:

defining a kernel for a given radius each time a second next item is selected after a baseline item is established by the user.

17. The computer program product of claim 15, wherein the at least one non-transitory computer readable medium further stores instructions translatable by the at least one processor to perform:

determining a conditional probability that the user will select a particular first next item after a particular baseline item is established by the user.

18. The computer program product of claim 15, wherein the at least one non-transitory computer readable medium further stores instructions translatable by the at least one processor to perform:

making a prediction as to which next item the user will select;

comparing the prediction with an actual next item selected by the user; and

assigning a penalty value to the prediction if the prediction is incorrect.

19. The computer program product of claim 18, wherein the prediction is one of a plurality of predictions and wherein the at least one non-transitory computer readable medium further stores instructions translatable by the at least one processor to perform:

determining a total penalty value for all incorrect predictions in the plurality of predictions; and

determining a set of weights that minimizes the total penalty.

20. A method for identifying consumer items more likely to be bought by a user, comprising:

in a computer having a processor and a memory, wherein the computer is communicatively connected to a computing device;

generating an interface which is presented to the user via the computing device;

determining, responsive to the user viewing a baseline item via the interface on the computing device, for each individual item of a plurality of items, a probability that the user will select the individual item, given that the user has expressed interest in the baseline item, wherein the probability is determined based on;

a similarity among the plurality of items based on observable features thereof;

an aggregate online search behavior relative to individual items across multiple users; and

preferences of the user with respect to the baseline item, wherein the baseline item is viewed by the user; and

ranking, in decreasing order, probabilities determined for the plurality of items, wherein items that are more likely to be bought by the user are ranked higher than items that are less likely to be bought by the user;

determining a probability that the user will select a particular one of the individual items of the plurality of items after the particular baseline item is viewed by the user; and

updating the interface running on the computing device, wherein the updated interface presents to the user at least a portion of the plurality of items ranked in decreasing order according to the probabilities;

after and in response to the user selecting one of the displayed items, ranking other ones of the portion of the plurality of items and updating the interface running on the computing device to present to the user the other ones of the portion of the plurality of items ranked in decreasing order according to the corresponding probabilities.