TU Darmstadt / ULB / TUbiblio

User Similarity and Deviation Analysis for Adaptive Visualizations

Nazemi, Kawa and Retz, Wilhelm and Kohlhammer, Jörn and Kuijper, Arjan (2014):
User Similarity and Deviation Analysis for Adaptive Visualizations.
Springer, Berlin, Heidelberg, New York, In: Human Interface and the Management of Information. Proceedings Part I, In: Lecture Notes in Computer Science (LNCS); 8521, DOI: 10.1007/978-3-319-07731-4₇,
[Conference or Workshop Item]

Abstract

Adaptive visualizations support users in information acquisition and exploration and therewith in human access of data. Their adaptation effect is often based on approaches that require the training by an expert. Further the effects often aim to support just the individual aptitudes. This paper introduces an approach for modeling a canonical user that makes the predefined training-files dispensable and enables an adaptation of visualizations for the majority of users. With the introduced user deviation algorithm, the behavior of individuals can be compared to the average user behavior represented in the canonical user model to identify behavioral anomalies. The further introduced similarity measurements allow to cluster similar deviated behavioral patterns as groups and provide them effective visual adaptations.

Item Type: Conference or Workshop Item
Erschienen: 2014
Creators: Nazemi, Kawa and Retz, Wilhelm and Kohlhammer, Jörn and Kuijper, Arjan
Title: User Similarity and Deviation Analysis for Adaptive Visualizations
Language: English
Abstract:

Adaptive visualizations support users in information acquisition and exploration and therewith in human access of data. Their adaptation effect is often based on approaches that require the training by an expert. Further the effects often aim to support just the individual aptitudes. This paper introduces an approach for modeling a canonical user that makes the predefined training-files dispensable and enables an adaptation of visualizations for the majority of users. With the introduced user deviation algorithm, the behavior of individuals can be compared to the average user behavior represented in the canonical user model to identify behavioral anomalies. The further introduced similarity measurements allow to cluster similar deviated behavioral patterns as groups and provide them effective visual adaptations.

Series Name: Lecture Notes in Computer Science (LNCS); 8521
Publisher: Springer, Berlin, Heidelberg, New York
Uncontrolled Keywords: Business Field: Visual decision support, Research Area: Human computer interaction (HCI), Adaptive information visualization, Machine learning, User modeling, Semantics visualization
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Interactive Graphics Systems
Event Title: Human Interface and the Management of Information. Proceedings Part I
Date Deposited: 12 Nov 2018 11:16
DOI: 10.1007/978-3-319-07731-4₇
Export:

Optionen (nur für Redakteure)

View Item View Item