Browse Prior Art Database

Method & System to Generate Comprehensible Rules for Real Time Personalization of Web Content

IP.com Disclosure Number: IPCOM000249479D
Publication Date: 2017-Feb-28
Document File: 4 page(s) / 167K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed are a system and method applied to web page personalization in website design for businesses that retains the user-friendliness of the rule-based approach and simultaneously takes advantage of behavioral and other data collected about the end users. This solution defines a comprehensible rule set of recommended rules that are presented to the marketer for use in website design.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 39% of the total text.

1

Method & System to Generate Comprehensible Rules for Real Time Personalization of Web Content

In designing the company website, many businesses prefer to have certain areas, called zones , (e.g., a banner at the top or bottom of the web site’s home page) present personalized content to the end user to achieve certain business goals. A business might want to show different content to different end users in order to maximize the business interest. Examples of business goals are to maximize the click rate of a zone, maximize the chance of the end user performing a certain target action (e.g., view a certain target page, complete a registration form, make a purchase in the same session, etc.). For each zone, the business pre-selects a number of alternative content variants, and a system determines which content variants to present when an end user lands on the web page containing the zone. The decision is made in real time, depending on the characteristics (called attributes) of the end user.

End user attributes can be of the behavioral type (e.g., has viewed a certain web page), or of the profile type (e.g., lives in a certain city). The two existing approaches to provide Real-Time Personalization (RTP) are:

 Rule-based approach: The business user who manages the RTP zone (e.g., a website designer or web marketer) manually specifies a set of if-then rules to control the choice of content variant when an end user lands on the zone. For example, if the end user comes on a Friday, then show content variant B otherwise show content variant A. This approach can be implemented using regular website development tools.

 Optimization approach: An RTP system controls the zone and uses a mathematical optimization engine that calculates the choice of a content variant to optimize the selected business goal, based on end user behavior data collected by the system thus far.

The rule-based approach is user-friendly. Rules and, hence, the zone behavior are completely under the control of the web marketer. The choice of content variant is clear to any business user reading the rules. The marketer’s knowledge can be translated into an if-then rule form in a straightforward way; hence, such a system is more easily accepted and used by a marketer. Different business objectives (e.g., maximize response rate, promote a new product) can be accommodated using the same if-then rule form.

To retain the user-friendliness of the rule-based approach and simultaneously take advantage of behavioral and other data collected about the end users, there is a need to have an systematic method to calculate recommended rules to the marketer. The recommended rules should be in the same form as that of a manually devised rule. The recommended rules should not be very complex so that the marketer can understand them without a lot of effort.

The novel contribution to knowledge is a comprehensible rule set (defined below) of recommended rules presented to the marketer. The marketer c...