Pangenetic web user behavior prediction system

• US 8,108,406 B2
• Filed: 12/30/2008
• Issued: 01/31/2012
• Est. Priority Date: 12/30/2008
• Status: Active Grant

• Current Assignee: 23andme Incorporated (23andMe Holding Co.)
• Original Assignee: Expanse Networks, Inc.
• Inventors: Kenedy, Andrew Alexander, Eldering, Charles Anthony
• Primary Examiner(s): TRUONG, CAM Y T
• Application Number:
  US12/346,738
• Publication Number:
  US 20100169343A1
• Time in Patent Office: 1,127 Days
• Field of Search: 707/749, 707/750, 707/758, 707/999.05, 707/999.06
• US Class Current: 707/749
• CPC Class Codes: G06F 16/2237 Vectors, bitmaps or matrices
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9538 Presentation of query results

Abstract

Computer based systems, methods, software and databases are presented in which correlations between web item preferences, behaviors and pangenetic (genetic and epigenetic) attributes of individuals are used for pangenetic based user behavior prediction in which predictions of a user'"'"'s online behavior can be generated based on the user'"'"'s pangenetic makeup. Data masking can be used to maintain privacy of sensitive portions of the pangenetic data.

Claims

1. A computer based method for predicting user behavior, comprising:

a) receiving at least one item preference of a user;

b) accessing a pangenetic data profile associated with the user;

c) accessing a dataset containing one or more non-pangenetic attributes associated with the at least one item preference of the user, wherein pangenetic data are correlated with the one or more non-pangenetic attributes and each non-pangenetic attribute indicates a user behavior, wherein the correlated pangenetic data are combinations of pangenetic data selected from pangenetic profiles associated with a group of individuals, wherein associations between the pangenetic data and the one or more non-pangenetic attributes contained in the dataset are previously determined based on statistical associations between non-pangenetic attributes and pangenetic data associated with the group of individuals, and wherein the correlations between the pangenetic data and the one or more non-pangenetic attributes contained in the dataset are determined by results of computing statistical associations which indicate strength of association between the non-pangenetic attributes and the pangenetic data associated with the group of individuals;

d) determining for said each non-pangenetic attribute, quantity of matches between the pangenetic data correlated with said each non-pangenetic attribute and the pangenetic data associated with the user; and

e) transmitting as output, based on the quantity of matches determined for said each non-pangenetic attribute, at least one non-pangenetic attribute to indicate at least one behavior predicted for the user, wherein the quantity of matches determined for said each non-pangenetic attribute is used to compute a pangenetic similarity value for said each non-pangenetic attribute, and the non-pangenetic attributes having pangenetic similarity values meeting a predetermined threshold value, and wherein the correlations between the pangenetic data and the one or more non-pangenetic attributes contained in the dataset comprise statistical associations indicating level of certainty, and wherein the level of certainty that the user will exhibit the predicted behavior is also transmitted as output.

2. The computer based method of claim 1, wherein the transmitting in step (e) is to at least one destination selected from the group consisting of the user, a database, a dataset, a computer readable memory, a computer readable medium, a computer processor, a computer network, a printout device, a visual display, and a wireless receiver.

3. The computer based method of claim 1, wherein the non-pangenetic attribute for which the largest quantity of matches is determined in step (d) is the at least one non-pangenetic attribute transmitted as output.

4. The computer based method of claim 1, further comprising computing a score for each non-pangenetic attribute using a quantitative similarity measure that processes the quantity of matches, and selecting the non-pangenetic attribute having the highest score for transmission as output in step (e).

5. The computer based method of claim 1, wherein the pangenetic data correlated with the one or more non-pangenetic attributes are pangenetic metadata.

6. The computer based method of claim 1, further comprising receiving one or more non-pangenetic attributes associated with the user, wherein the one or more non-pangenetic attributes associated with the user match one or more non-pangenetic attributes associated with the group of individuals.

7. The computer based method of claim 1, further comprising acts of

i) transmitting an authorization request for access to the pangenetic data associated with the user, and

ii) receiving an authorization granting access to the pangenetic data associated with the user.

8. The computer based method of claim 7, wherein the transmitting in act (i) is to a mobile computing device, and wherein the receiving in act (ii) is from the mobile computing device.

9. The computer based method of claim 1, wherein the at least one non-pangenetic attribute transmitted as output in step (e) is used for a task selected from the group consisting of selecting data for retrieval, selecting data for visual display, selecting the locations of data in a visual display, formulating a internet search query, and selecting web based items for recommendation to a user.

10. The computer based method of claim 1, wherein the dataset is an item feedback matrix.

11. The computer based method of claim 10, further comprising:

i) receiving non-pangenetic attribute data associated with a group of individuals, wherein the non-pangenetic attribute data indicate behaviors of the individuals with respect to the at least one item preference;

ii) accessing pangenetic data associated with the individuals;

iii) determining correlations between the non-pangenetic attribute data and the pangenetic data associated with the individuals; and

iv) storing the correlations between the non-pangenetic attribute data and the pangenetic data to generate an item feedback matrix.

12. The computer based method of claim 1, wherein the accessing in step (b) is in accordance with an applied data mask, the method further comprising:

i) transmitting an authorization request for access to the pangenetic data associated with the user;

ii) receiving an authorization which grants access to the pangenetic data;

iii) accessing a data mask, wherein the data mask'"'"'s parameters are associated with the authorization; and

iv) applying the data mask to the pangenetic data.

13. A program storage memory readable by a machine and containing a set of instructions which, when read by the machine, causes execution of a computer based method for predicting user behavior, comprising:

a) receiving at least one item preference of a user;

b) accessing pangenetic data associated with the user;

c) accessing a dataset containing one or more non-pangenetic attributes associated with the at least one item preference of the user, wherein the pangenetic data are correlated with the one or more non-pangenetic attributes and each non-pangenetic attribute indicates a user behavior , wherein the correlated pangenetic data are combinations of pangenetic data selected from pangenetic profiles associated with a group of individuals, wherein associations between the pangenetic data and the one or more non-pangenetic attributes contained in the dataset are previously determined based on statistical associations between non-pangenetic attributes and pangenetic data associated with the group of individuals, and wherein the correlations between the pangenetic data and the one or more non-pangenetic attributes contained in the dataset are determined by results of computing statistical associations which indicate strength of association between the non-pangenetic attributes and the pangenetic data associated with the group of individuals;

d) determining for said each non-pangenetic attribute, the quantity of matches between the pangenetic data correlated with said each non-pangenetic attribute and the pangenetic data associated with the user; and

e) transmitting as output, based on the quantity of matches determined for said each non-pangenetic attribute, at least one non-pangenetic attribute to indicate at least one behavior predicted for the user, wherein the quantity of matches determined for said each non-pangenetic attribute is used to compute a pangenetic similarity value for said each non-pangenetic attribute, and the non-pangenetic attributes having pangenetic similarity values meeting a predetermined threshold value, and wherein the correlations between the pangenetic data and the one or more non-pangenetic attributes contained in the dataset comprise statistical associations indicating level of certainty, and wherein the level of certainty that the user will exhibit the predicted behavior is also transmitted as output.

14. The program storage memory of claim 13, wherein the transmitting in step (e) is to at least one destination selected from the group consisting of the user, a database, a dataset, a computer readable memory, a computer readable medium, a computer processor, a computer network, a printout device, a visual display, and a wireless receiver.

15. A computer database system for predicting user behavior, comprising:

a) a memory containing;

i) a first data structure containing pangenetic data associated with a user;

ii) a second data structure containing one or more non-pangenetic attributes associated with at least one item preference of the user, wherein the pangenetic data are correlated with the one or more non-pangenetic attributes and each non-pangenetic attribute indicates a user behavior , wherein the correlated pangenetic data are combinations of pangenetic data selected from pangenetic profiles associated with a group of individuals, wherein associations between the pangenetic data and the one or more non-pangenetic attributes contained in the dataset are previously determined based on statistical associations between non-pangenetic attributes and the pangenetic data associated with the group of individuals, and wherein correlations between the pangenetic data and the one or more non-pangenetic attributes contained in the dataset are determined by results of computing statistical associations which indicate strength of association between the non-pangenetic attributes and the pangenetic data associated with the group of individuals ;

b) a processor for;

i) receiving the at least one item preference associated with the user;

ii) accessing the first data structure;

iii) accessing the second data structure;

iv) determining for said each non-pangenetic attribute, quantity of matches between the pangenetic data correlated with said each non-pangenetic attribute and the pangenetic data associated with the user; and

v) transmitting as output, based on the quantity of matches determined for said each non-pangenetic attribute, at least one non-pangenetic attribute to indicate at least one behavior predicted for the user, wherein the quantity of matches determined for said each non-pangenetic attribute is used to compute a pangenetic similarity value for said each non-pangenetic attribute, and the non-pangenetic attributes having pangenetic similarity values meeting a predetermined threshold value, and wherein the correlations between the pangenetic data and the one or more non-pangenetic attributes contained in the dataset comprise statistical associations indicating level of certainty, and wherein the level of certainty that the user will exhibit the predicted behavior is also transmitted as output.

16. The computer database system of claim 15, wherein the transmitting in part (v) is to at least one destination selected from the group consisting of the user, a database, a dataset, a computer readable memory, a computer readable medium, a computer processor, a computer network, a printout device, a visual display, and a wireless receiver.