Information system with propensity modelling and profiling engine

• US 20070050201A1
• Filed: 05/25/2006
• Published: 03/01/2007
• Est. Priority Date: 05/26/2005
• Status: Abandoned Application

• Current Assignee: MoneyExpert Ltd.
• Original Assignee: MoneyExpert Ltd.
• Inventors: O'Shaughnessy, Alex, Gardner, Sean
• Application Number:
  US11/441,885
• Publication Number:
  US 20070050201A1
• Time in Patent Office: Days
• US Class Current: 705/1.100
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0282 Rating or review of business operators or products

Abstract

An information system having various databases (30-40 and 50-52) for storing data relating to products or services, and risk data models or propensity models. A web server (48) provides web pages (20) through which a user is able to express choices, which result in querying the product database (32) to provide a raw result set, and filtering of the raw result set based on a risk model to eliminate results not having a likelihood of a successful outcome. The filtered result set is displayed. A propensity model database may be provided for storing data modelling the propensity of an outcome in relation to related products or services, e.g. based upon demographic or lifestyle data. A tracking tool may be used to define past behaviour of the user.

Claims

1. An information system comprising:

a product database for storing data relating to products or services,

a risk model database for storing data modelling the likelihood of an outcome in relation to the products or services, and

a web server coupled to the product database and the risk model database, wherein the web server comprises;

means for providing web pages through which a user is able to express choices,

query means for querying the product database in accordance with the user choices to provide a raw result set,

filter means for providing a filtered result set by filtering the raw result set based on the risk model database to eliminate results not having a likelihood of a successful outcome, and

display means for displaying the filtered result set.

2. An information system in accordance with claim 1, wherein the risk model database scores likelihood of a successful outcome based upon at least one of:

postcode, demographic data and lifestyle data.

3. An information system in accordance with claim 1, wherein the risk model data is built up using tracking data defining past behaviour of the user when using the website.

4. An information system in accordance with any claim 1, further comprising:

a risk profiling database coupled to the web server, wherein the web server provides means to invite a user to enter user data and wherein the web server queries the risk profiling database using the user data to thereby deliver at least one risk score;

means for obtaining a preferred range of risk scores for each result within the result set; and

refining means for refining the result set based upon the risk score for the user,

wherein the display means displays the refined results.

5. An information system in accordance with claim 4, wherein the user data comprises at least one of:

user identity and postcode/zipcode.

6. An information system in accordance with any claim 1, further comprising means for inviting a user to express preferred attributes, means for capturing a preferred attribute set for the user, and means for rank ordering the filtered result set based on the preferred attribute set.

7. An information system in accordance with claim 6, wherein the means for rank ordering takes into account the user'"'"'s preferred attribute set relative to preferred attributes previously obtained through sampling.

8. An information system in accordance with claim 1, further comprising a customer profiling engine for developing a profile for a user, wherein the web server provides the user with a click-through button for each of the displayed results to direct the user to a provider website at which the corresponding product or service can be obtained, and wherein the website captures data relating to a click-through action by the user and directs this data to the customer profiling engine to update a profile for that user.

9. An information system in accordance with claim 8, wherein the web server has an interface with a provider server that serves the provider website, and wherein the web server delivers to the provider server a universal resource locator (URL) having profile data for the user appended thereto.

10. An information system in accordance with claim 1, further comprising means for delivering an applet or script to a browser of a user to capture user actions while the browser is browsing web pages served by the web server, and wherein the web server is arranged to maintain a session with the browser and to obtain the user actions from the applet or script during the session.

11. An information system in accordance with claim 10, wherein the web server is arranged to obtain from the user'"'"'s browser information indicating that the user has returned to the web pages served by the web server following a browsing sequence in which web pages of another server have been browsed.

12. An information system in accordance with claim 1, further comprising a propensity model database for storing data modelling the propensity of an outcome in relation to related products or services.

13. An information system in accordance with claim 12, wherein the propensity model database scores propensity towards a successful outcome based upon at least one of:

demographic data and lifestyle data.

14. An information system in accordance with claim 12, wherein the propensity model data is built up using tracking data defining past behaviour of the user when using the website.

15. An information system comprising:

a web server for serving web pages to an internet, the web server having means for delivering an applet or script to a browser of a user to capture user actions and means for retrieving the user actions from the applet or script, and

a profiling engine for performing correlations on actions of users to thereby categorize users according to browsing actions.

16. An information system in accordance with claim 15, wherein the profiling engine correlates actions of users with profile data for those users to search for correlations therebetween.

17. An information system in accordance with claim 15, wherein the user actions include user selections among choices presented by the web pages.

18. An information system in accordance with claim 15, comprising means for comparing user actions with stated user preferences, means for identifying user actions that are contradictory to stated user preferences and means for adjusting a user profile when a user takes an action that is contradictory to a stated preference.

21. An information system in accordance with claim 15, wherein the web pages invite the user to select products or services having different brand categories, and wherein the user actions include selection of products or services by brand category.

19. An information system comprising a web server for serving web pages to an internet, the web server having means for delivering an applet or script to a browser of a user to capture user actions and means for retrieving the user actions from the applet or script, and

a profiling database having behaviour definitions stored therein, and having means for comparing captured user actions with behaviour definitions to categorize a user according to the user actions.

20. An information system in accordance with claim 19, wherein the user actions include user selections among choices presented by the web pages.

22. An information system in accordance with claim 19, comprising means for comparing user actions with stated user preferences, means for identifying user actions that are contradictory to stated user preferences and means for adjusting a user profile when a user takes an action that is contradictory to a stated preference.

23. An information system comprising means for receiving from a third party historical data describing the outcome of previous referrals, and means for refining a risk model using the historical data, wherein the risk model is used to control a filter to provide a filtered result set by filtering the raw result set based on the risk model database to eliminate selected results.