User-Adapted Information Services

User-Adapted Information Services

Thomas Mandl (University of Hildesheim, Germany)
DOI: 10.4018/978-1-59904-879-6.ch034
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

This chapter describes personalization strategies adopted in digital libraries. Personalization and individualization are introduced as means to improve the usability of digital library services. The goal of personalization for digital libraries is mainly the presentation of individual results to the user. This can be modelled based on a user interest model which is applied during the search process. Two users with the same query can receive different results based on their interest profile maintained by the system. Typical approaches and systems for individualizing the results of information retrieval systems are presented. The retrieval process is described. Knowledge sources and common knowledge representation for personalization are elaborated. Most common, the search history and documents accessed in the past are exploited for modelling the user interest. Finally, the chapter mentions drawbacks and success factors for personalization and individualization systems.
Chapter Preview
Top

Personalization And Individualization

Users are Different

Users can differ in many ways like age or culture. Most important for knowledge work which is supported by information system are cognitive differences. Users may differ in their knowledge about the interaction with the system and the domain. The differences between beginners and advanced users are an issue which is often exploited for personalization. At the same time, users may be different according to their knowledge (e.g., in a e-learning system, they may have reached different levels of knowledge). Resistance to change, intelligence, intro-/extroversion, fear of failure, and creativity are further personal features. For some interfaces, the spatial orientation capabilities may be of importance. Users can have different preferences in interaction styles. Some may prefer the keyboard, others the mouse, and again others, spoken language. Obviously, adaptation does contradict the human-computer interaction principle of consistency for interfaces. As a consequence, adaptation needs to improve the system up to an extent which exceeds these potential shortcomings. Certainly, adaptation is of specific value for beginners who start to use a system. However, for this group, it is hard to acquire knowledge.

Top

Modifying A System

The modification of the system can affect the user interface (functions, appearance, way of interaction), the content (different knowledge objects), or the presentation (sequence, level of detail). For a digital library all of these three aspects can be of interest, however, content adaptation has been attracted most research in the digital library community. Content adaptation will be thoroughly discussed in the following section. Typical content adaptation is implemented by recommender systems which suggest new content items to users in e-commerce applications.

User interface adaptation has been the focus of much research. The adaptation initiated by users has been integrated in many systems for many years. Menus, tool bars, and other aspects of graphical user interfaces can be changed by users. Some aspects like desktop background pictures or ring tones for mobile phones are heavily used to express individuality through aesthetic elements.

Key Terms in this Chapter

Association Rules: Association rules describe relationships and correlations between attributes or objects in large data sets. Several algorithms have been developed to extract such rules from large data sets.

User Model: The user model is the collection of knowledge and assumption of the system about one user.

Machine Learning: Machine learning is a subfield of Artificial Intelligence which provides algorithms for the discovery of relations or rules in large data sets. Machine learning leads to functions which can automatically classify or categorize objects based on their features. Inductive learning from labeled examples is the most well known application.

Information Retrieval: Information retrieval is concerned with the representation and knowledge and subsequent search for relevant information within these knowledge sources. Information retrieval provides the technology behind search engines.

Human-Computer Interaction (HCI): Deals with the optimization of interfaces between human users and computing systems. Technology needs to be adapted to the properties and the needs of users. The knowledge sources available for this endeavor are guidelines, rules, standards, and results from psychological research on the human perception and cognitive capabilities. Evaluation is necessary to validate the success of interfaces.

Smart Home: Enriching houses with ubiquitous technology can lead to a smart home which better supports its inhabitants by automatically regulating its functions.

Adaptation: Adaptation can be seen as a process of modification based on input or observation. An information system should adapt itself to the specific needs of individual users.

Complete Chapter List

Search this Book:
Reset