Authors:
Muhammad Faisal Cheema
1
;
Stefan Jänicke
1
;
Judith Blumenstein
2
and
Gerik Scheuermann
1
Affiliations:
1
Institute for Computer Science and Leipzig University, Germany
;
2
Leipzig University, Germany
Keyword(s):
Text Visualization, Digital Humanities, Distant Reading, Concept Search, Information Retrieval, Concept Editor, Concept Modeling.
Related
Ontology
Subjects/Areas/Topics:
Abstract Data Visualization
;
Computer Vision, Visualization and Computer Graphics
;
General Data Visualization
;
Graph Visualization
;
Information and Scientific Visualization
;
Text and Document Visualization
;
Visual Data Analysis and Knowledge Discovery
;
Visualization Applications
;
Visualization Taxonomies and Models
Abstract:
We introduce a concept search environment that caters for the needs of humanities scholars who want to improve
the accuracy of search results when querying historical text corpora. For this purpose, we designed a
so-called Concept Editor that allows to model historical concepts in a diagram style according to the imaginations
of the humanities scholar. For the inspection of results determined in the proposed concept search,
we provide a Concept Search Results Viewer that uses the existent layout of the underlying concept model
to visualize related texts according to the relevance to the given concept. We further designed the overall
system the way that the humanities scholar can iteratively refine the concept idea, which leads to a gradual
improvement of search results. To illustrate the whole development pipeline, we provide a usage scenario
on modeling the concept epilepsy with the purpose of improving the accuracy of results compared to usual
applied keyword-based search methods.