User Models for Adaptive Information Retrieval on the Web: Towards an Interoperable and Semantic Model

User Models for Adaptive Information Retrieval on the Web: Towards an Interoperable and Semantic Model

Max Chevalier (Université Toulouse 3 Paul Sabatier, IRIT, UMR 5505, France), Christine Julien (Université Toulouse 3 Paul Sabatier, IRIT, UMR 5505, France) and Chantal Soulé-Dupuy (Université Toulouse 1 Capitole, IRIT, UMR 5505, France)
DOI: 10.4018/jaras.2012070101
OnDemand PDF Download:
No Current Special Offers


Searching information can be realized thanks to specific tools called Information Retrieval Systems IRS (also called “search engines”). To provide more accurate results to users, most of such systems offer personalization features. To do this, each system models a user in order to adapt search results that will be displayed. In a multi-application context (e.g., when using several search engines for a unique query), personalization techniques can be considered as limited because the user model (also called profile) is incomplete since it does not exploit actions/queries coming from other search engines. So, sharing user models between several search engines is a challenge in order to provide more efficient personalization techniques. A semantic architecture for user profile interoperability is proposed to reach this goal. This architecture is also important because it can be used in many other contexts to share various resources models, for instance a document model, between applications. It is also ensuring the possibility for every system to keep its own representation of each resource while providing a solution to easily share it.
Article Preview

User Modeling In The Domain Of Information Retrieval

The role of the user is very important at different steps of the information retrieval process (querying, analyzing search results …). Moreover, as users’ satisfaction is the main goal of information retrieval systems, describing the users precisely is a key step to reach this goal. So, a user model (also called profile) is required to achieve an adaptive information retrieval (AIR) process. Such a model allows the system to better know the user, his information needs and his behavior (Finin, 1989). In a general point of view, it can be composed of data and rules. In the specific Information Retrieval field, a user model is rather commonly described by a set of eventually weighted characteristics that describe the user himself (Korfhage, 1997).

In this paper user modeling is studied under two main aspects: model definition and model content. It does not concern the way the user models are exploited and the way they evolve along time. Note that the discussion about user models is the same for either individual user modeling or user group modeling.

User Model Definition

User modeling implies the identification or the elicitation of features (data, rules …) that characterize each user in a specific context or for the realization of a particular task. Several techniques exist and two main approaches can be identified in the AIR field to define user models. Thus, stereotypes or profiling techniques can be applied to associate to a user a model that characterizes him.

  • Profiling (Cho et al., 2002) consists in tracking the user during his different log sessions and in analyzing his behavior. Every user is associated to a personal model that characterizes him;

  • Stereotypes (Shapira et al., 1997) consist in associating users to pre-defined classes. These classes contain specific characteristics and values (i.e., a stereotype for students). The stereotypes approach is a kind of clustering and is mainly used for defining users’ groups or communities.

Complete Article List

Search this Journal:
Open Access Articles: Forthcoming
Volume 7: 1 Issue (2016)
Volume 6: 2 Issues (2015)
Volume 5: 4 Issues (2014)
Volume 4: 4 Issues (2013)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing