How to visualize a crisp or fuzzy topic set over a taxonomy

Authors:
Boris Mirkin;Susana Nascimento;Trevor Fenner;Rui Felizardo
Affiliations:
Division of Applied Mathematics and Informatics, National Research University, Higher School of Economics, Moscow, Russian Federation and Department of Computer Science, Birkbeck University of Lon ...;Department of Computer Science and Centre for Artificial Intelligence, Universidade Nova de Lisboa, Caparica, Portugal;Department of Computer Science, Birkbeck University of London, London, UK;Department of Computer Science and Centre for Artificial Intelligence, Universidade Nova de Lisboa, Caparica, Portugal
Venue:
PReMI'11 Proceedings of the 4th international conference on Pattern recognition and machine intelligence
Year:
2011

Citing 4
Cited 0

Introduction to Information Visualization

Introduction to Information Visualization
Seeing the forest for the trees

Bioinformatics
Using ontologies to facilitate post-processing of association rules by domain experts

Information Sciences: an International Journal
Advanced processing for ontological queries

Proceedings of the VLDB Endowment

Quantified Score

Hi-index	0.01

Visualization

Abstract

A novel method for visualization of a fuzzy or crisp topic set is developed. The method maps the set's topics to higher ranks of the taxonomy tree of the field. The method involves a penalty function summing penalties for the chosen "head subjects" together with penalties for emerging "gaps" and "offshoots". The method finds a mapping minimizing the penalty function in recursive steps involving two different scenarios, that of 'gaining a head subject' and that of 'not gaining a head subject'. We illustrate the method by applying it to illustrative and real-world data.