Portable Personality and its Personalization Algorithms: An Overview and Directions

Portable Personality and its Personalization Algorithms: An Overview and Directions

Stefan Uhlmann (Tampere University of Technology, Finland) and Artur Lugmayr (Tampere University of Technology, Finland)
Copyright: © 2012 |Pages: 28
DOI: 10.4018/978-1-60960-774-6.ch004
OnDemand PDF Download:
No Current Special Offers


With the advances in ubiquitous computing, there is an increasing focus on personalization of user information especially in web-based applications and services. Currently those personalized user profiles are scattered, mostly stored for each individual service. Therefore, this prohibits the usage of those profiles in different environments such as other web-based services, shopping in local stores or sharing interests among people. The so-called Portable Personality focuses on the management and distribution of personalized profiles (in form of a digital personality representing the real-world user) through mobile devices. These portability aspects merge with the idea of cross-system personalization using a single generic user profile. We will briefly introduce some aspects related to profile representation and management with focus on attempts towards such a generic representation. The main discussion will be concentrated around profile portability and its effects on personalization especially towards cross-system support. We include different portable profile scenarios and their personalization methodologies. Furthermore, current personalization algorithms are considered with possible associations towards the presented portable scenarios. At the end, we reflect on existing challenges of current approaches in the field of portable personalization and try to provide some recommendations.
Chapter Preview

User Profiles And Their Management

Personalization of any kind of information is evolving in a rapid manner especially for web-based applications and services whether in advertisement, search engines, online shopping, or social networks. Hence, the collection and application of personalized information is currently omnipresent. We understand personalization as tailoring and providing content and services to individuals and groups based on knowledge about their preferences and behavior. This can range from simple superficial factors such as custom ring-tones to the complex tailoring of the presentation of a shopping web site to a user's personal interests and their previous purchasing behavior. To make use of such information, a so-called ‘personalized user profile’ (UP) is to be generated preferably without user intervention. Such user profile may vary from a rather simple to complex representations depending on how much and what type of information is gathered and stored. A simple relation is pictured in Figure 1. An example what kind of information might be stored in a rather complex user profile is illustrated in Figure 2. (Note that in the annual personalization survey from www.choicestream.com in 2009, personalization and recommendations are well received and considered useful to make purchases. However, they also found that the quality of recommendations decreased to previous year 2008 as well as recommendations can widely vary depending on different retail categories.)

Figure 1.

Complexity of user profiles

Figure 2.

Information stored in a complex user profile


The general concept is to gather user-specific information about the user (profiling), to manage and store this information (content management), distribute it to consumer applications or services (profile distribution, portability), and finally extracting those pieces of information valuable to the consumer current needs (profile evaluation, personalization).

Current UPs are mainly used in web applications to personalize searches, advertisements and shopping recommendations such as music, movies, and books. Most of the time, the user has a different profile for every online shop, service, or website such as last.fm (http://www.dataportability.org) tries to develop a standard that allows users to gain control over their own data again.

Complete Chapter List

Search this Book: