Method and System for Content Based Demographics Prediction for Websites

• US 20100161385A1
• Filed: 12/21/2009
• Published: 06/24/2010
• Est. Priority Date: 12/19/2008
• Status: Abandoned Application

• Current Assignee: IPwe, Inc.
• Original Assignee: nXn Tech LLC
• Inventors: Karypis, George, Han, Eui-Hong
• Application Number:
  US12/643,904
• Publication Number:
  US 20100161385A1
• Time in Patent Office: Days
• US Class Current: 705/10
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities
• CPC Class Codes: G06Q 30/0254 based on statistics

Abstract

Systems and methods for predicting characteristics of a web user, determining a combination of websites to obtain a target demographic mix, determining a set of keywords to buy to obtain a target demographic mix, selecting websites from market research and designing websites to appeal to an audience with desired demographic characteristics. Systems and methods may include determining features of web-pages of ad-carrying, target websites, applying prediction models to the determined features of the ad-carrying, target websites to predict values of demographic attributes of the ad-carrying, target websites, receiving one or more inputs including a target demographic mix, receiving a number that indicates an amount of visitors of the ad-carrying, target websites, and determining a combination of websites that provide target demographic mix based on the predicted values of the demographic attributes and number of visitors of the ad-carrying, target websites.

Claims

1. A method of predicting characteristics of a web user comprising:

receiving online session browsing history of a user, wherein browsing history identifies websites visited by a user during one or more online sessions, wherein identified websites include known websites and unknown websites;

retrieving known demographic attributes data of known websites included in identified websites, wherein the known demographic attributes data includes known values of demographic attributes of the known websites;

determining features of web-pages of unknown websites; and

applying prediction models to the determined features of the unknown websites to predict values of unknown demographic attributes of the unknown websites.

2. The method of claim 1 further comprising combining the known values of the known demographic attributes and the predicted values of the unknown demographic attributes.

3. The method of claim 2 wherein the combining combines the known values and the predicted values to predict a distribution of demographic attributes of the user.

4. The method of claim 3 further comprising:

receiving updated browsing history of the user; and

updating the predicted distribution based on the received updated browsing history of the user.

5. A computer-readable medium comprising instructions stored thereon that may be executed by a computer for performing the method of claim 1.

6. A method of determining a combination of websites to obtain a target demographic mix, comprising:

determining features of web-pages of target websites;

applying prediction models to the determined features of the ad-carrying, target websites to predict values of demographic attributes of the target websites;

receiving one or more inputs including a target demographic mix;

receiving a number that indicates an amount of visitors of the target websites; and

determining a combination of websites that provide target demographic mix based on the predicted values of the demographic attributes and number of visitors of the target websites.

7. The method of claim 6 wherein the determining a combination of websites processes the predicted values of the demographic attributes and number of visitors of the target websites using an optimization method.

8. The method of claim 6 further comprising receiving one or more secondary objectives.

9. The method of claim 8 wherein the one or more secondary objectives include one or more secondary objectives chosen from a list consisting of:

total advertising cost, total time that is required to reach audience with the target demographic mix, and a number of ads that may be placed.

10. The method, of claim 8 wherein the determining a combination of websites processes the predicted values of the demographic attributes, the number of visitors of the ad-carrying, target websites and the one or more secondary objectives using an optimization method.

11. The method of claim 8 wherein the one or more secondary objectives include:

total advertising cost, total time that is required to reach a forecasted target audience, prior partnership information, and competitor information.

12. A computer-readable medium comprising instructions stored thereon that may be executed by a computer for performing the method of claim 6.

13. A method of determining a set of keywords to buy to obtain a target demographic mix comprising:

receiving one or more inputs including a target demographic mix;

identifying one or more sets of website combinations to reach the target demographic mix, wherein the identifying identifies the one or more sets of website combinations using extracted features of web-pages to predict demographic attribute values; and

analyzing a subset of web-pages of the one or more sets of website combinations to determine a set of terms that occur in the web-pages.

14. The method of claim 13 wherein the analyzing utilizes feature selection and optimization methods to determine the set of terms.

15. The method of claim 14 further comprising receiving one or more objectives.

16. The method of claim 15 wherein the feature selection and optimization methods optimize the set of terms to best meet the one or more objectives

17. The method of claim 15 further comprising weighting scores given to determined terms to give higher importance to one or more objectives.

18. The method of claim 13 wherein the identifying one or more sets of website combinations to reach the target demographic mix comprises:

determining features of web-pages of target websites;

applying prediction models to the determined features of the target websites to predict values of demographic attributes of the target websites;

receiving one or more inputs including a target demographic mix;

receiving a number that indicates an amount of visitors of the target websites; and

determining a combination of websites that provide target demographic mix based on the predicted values of the demographic attributes and number of visitors of the target websites.

19. A computer-readable medium comprising instructions stored thereon that may be executed by a computer for performing the method of claim 13.

20. A method of selecting websites from market research, comprising:

receiving one or more inputs including a target demographic mix;

determining features of web-pages of target websites;

applying prediction models to the determined features of the target websites to predict values of demographic attributes of the target websites;

receiving a number that indicates an amount of visitors of the target websites; and

determining a combination of websites that provide target demographic mix based on the predicted values of the demographic attributes and number of visitors of the target websites.

21. The method of claim 20 further comprising:

monitoring results of a market research campaign; and

re-optimizing the determined combination of websites based on the monitored results.

22. A computer-readable medium comprising instructions stored thereon that may be executed by a computer for performing the method of claim 20.

23. A method of designing websites to appeal to an audience with desired demographic characteristics, comprising:

receiving set of desired values for one or more demographic attributes;

determining a correlation between one or more features of web-pages and set of training websites that have desired values for the one or more demographic attributes, wherein the determining a correlations includes

identifying combination of features that would result in a prediction of the desired values.

24. The method of claim 23 further comprising designing a website to include the identified combination of features.

25. The method of claim 23 wherein the determining includes analyzing a prediction model developed using features extracted from a subset of web-pages of the training websites and obtained demographic attributes data of the training websites, wherein the prediction models may be used to predict one or more values for target demographic attributes.

26. A computer-readable medium comprising instructions stored thereon that may be executed by a computer for performing the method of claim 23.