Personalization is the ability to dynamically alter site behavior to appeal to an individual
visitor, based on the evaluation of a set of criteria.
Personalization Center provides a combination of fundamental and advanced personalization capabilities to
users of a Content Server/Content Center system.
|
Feature
|
Benefit
|
|
Business Manager |
|
Browser based administration UI |
An easy-to-use interface makes it possible for business managers to control site behaviour, site design and functionality without requiring IT intervention. This improves productivity of both business managers and IT professionals, minimizes bottlenecks and lowers the total cost of ownership of the content system. Personalization Center utilizes Content Server and Content Center admin and editorial screens to provide a familiar, consistent UI experience. |
|
Segment Creation and Management |
Segment is a new asset type unique to Personalization Center, managed in a similar fashion to existing Content Center assets. Business managers create Segments to place site visitors in more specific target groups. Personalization Center then associates rules to segments to refine the relevancy of displayed assets. |
|
Embedded Rules Engine |
Personalization Center features a Java-based rules engine that enables quick parsing of XML based rules. A rule is an If/Then/Else decision statement that determines the content to be displayed to a particular visitor based on their profile information, the segment they are associated with, and the business intelligence of the marketing manager. |
|
Multiple personalization techniques
- Concept Mapping (Autonomy)
|
Personalization in general can be divided into 2 basic methods - implicit ( the ability to derive an understanding of the visitor based on their behavior i.e clickstream) and explicit (the ability to understand a visitor based on specific information provided by the visitor). No single personalization technique is sufficient to provide a truly relevant user experience. Personalization Center offers the ability to combine aspects of both techniques and provide recommendations utilizing the following approaches;
- Visitors identify themselves in a variety of fashions. They can login (or register), thereby providing explicit information about themselves, their preferences and priorities or the system can gather clickstream and HTTP header information and implicitly infer such things as browser type, geographic location, season etc. Either technique is a relatively straightforward means of personalizing the visitor experience
- Rules can be defined by the Business Manager to maximize the interaction with the site visitor. Constructed in the familiar Content Center interface, rules will be invoked when keywords extracted from a visitor profile match a defined segment. Because the rules engine is Java based, rules based recommendation is faster than most "black box" techniques.
- The process of associating a visitor with similar visitors (e.g. people who bought this book also bought books by these 3 authors) is achieved through the use of a third party tool such as LikeMinds or NetPerceptions.
- A sophisticated method for providing related information utilizing 'concept signatures'. Personalization Center OEM's a subset of the Autonomy Knowledge Server to provide this capability. Typically used to produce related links between articles, it can also be used to derive related links between users (ie the formation of communities based on similar signatures). |
|
Search |
Also provided by virtue of the Autonomy OEM, allows visitors to navigate easily through sites with simple to use, natural language search. Utilizes the Autonomy concept signature to allow search by concept (in addition to simple text) meaning a more relevant result is returned. |
|
Personal channels |
Visitor can control his own experience by creating personalized "channels" of content. |
|
Content Server/Content Center |
|
Open Platform Architecture |
By virtue of running on top of Content Server/Content Center, Personalization Center takes advantage of the performance, scalability and reliability delivered by any of the 3 leading app servers ie WebLogic, Websphere and iPlanet. Code is written in Java and XML and leverages Content Server's adherence to J2EE standards. |
|
Security |
Personalization Center utilizes ACL's to restrict access to any of the functionality associated with updating visitor information. Any user familiar the use of ACL's in Content Center or Server will find this consistent. |
|
Authentication |
Personalization Center supports either the use of native Content Server authentication or (with some minimal customization) the use of LDAP. Either method can be used to facilitate the user registration process. |