Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Semantic Models for Adaptive Interactive SystemsPASTREM: Proactive Ontology Based Recommendations for Information Workers

Semantic Models for Adaptive Interactive Systems: PASTREM: Proactive Ontology Based... [Information work involves the frequent (re)use of information objects (e.g. files, web sites, emails) for different tasks. Information reuse is complicated by the scattered organization of information among different locations. Therefore, access support based on recommendations is beneficial. Still, support needs to consider the ad-hoc nature of information work and the resulting uncertainty of information requirements. We present PASTREM, an ontology-based recommender system which proactively proposes information objects for reuse while a user interacts with a computer. PASTREM reflects the ad-hoc nature of information work and allows users to switch seamlessly between recommendations for more multitasking oriented or more focused work. This chapter describes the PASTREM recommender, the used data foundation of interaction histories, data storage in an ontology and the process of recommendation elicitation. PASTREM is evaluated in comparison with other, activity related recommendation approaches for information reuse, namely last recently used, most often used, longest used and semantically related. We report on strength and weaknesses of the approaches and show the benefits of PASTREM as recommender which considers the difference between single task focused and multitasking oriented recommendations.] http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png

Semantic Models for Adaptive Interactive SystemsPASTREM: Proactive Ontology Based Recommendations for Information Workers

Part of the Human–Computer Interaction Series Book Series
Editors: Hussein, Tim; Paulheim, Heiko; Lukosch, Stephan; Ziegler, Jürgen; Calvary, Gaëlle
Schmidt, Benedikt; Godehardt, Eicke; Paulheim, Heiko
Semantic Models for Adaptive Interactive Systems — May 14, 2013
Download PDF
24 pages

Loading next page...
 
/lp/springer-journals/semantic-models-for-adaptive-interactive-systems-pastrem-proactive-zbD2KRMBWa
Publisher
Springer London
Copyright
© Springer-Verlag London 2013
ISBN
978-1-4471-5300-9
Pages
109 –132
DOI
10.1007/978-1-4471-5301-6_6
Publisher site
See Chapter on Publisher Site

Abstract

[Information work involves the frequent (re)use of information objects (e.g. files, web sites, emails) for different tasks. Information reuse is complicated by the scattered organization of information among different locations. Therefore, access support based on recommendations is beneficial. Still, support needs to consider the ad-hoc nature of information work and the resulting uncertainty of information requirements. We present PASTREM, an ontology-based recommender system which proactively proposes information objects for reuse while a user interacts with a computer. PASTREM reflects the ad-hoc nature of information work and allows users to switch seamlessly between recommendations for more multitasking oriented or more focused work. This chapter describes the PASTREM recommender, the used data foundation of interaction histories, data storage in an ontology and the process of recommendation elicitation. PASTREM is evaluated in comparison with other, activity related recommendation approaches for information reuse, namely last recently used, most often used, longest used and semantically related. We report on strength and weaknesses of the approaches and show the benefits of PASTREM as recommender which considers the difference between single task focused and multitasking oriented recommendations.]

Published: May 14, 2013

Keywords: Recommender System; Information Requirement; Latent Dirichlet Allocation; Information Object; Task Switch

There are no references for this article.