Visualizing a Thinker's Life.

Area of Science:

• Digital Humanities
• Information Visualization
• Literary Studies

Background:

• Analyzing medium-sized text collections, especially author-specific works, presents challenges in understanding structure and content.
• Existing methods may not adequately capture the nuances of an author's textual corpus, including chronological and entity-based relationships.

Purpose of the Study:

• To present a novel visualization framework designed for the analysis of medium-sized text collections.
• To introduce specific visualization techniques, TextGenetics and ConceptCircuit, for exploring authorial works.
• To enable complex querying of textual data through a wildcard-based phrase search.

Main Methods:

• Development of document-based visualization techniques: TextGenetics for structure and chronology, and ConceptCircuit for augmenting with key entities (persons, locations).
• Integration of a wildcard-based phrase search functionality into the visualization framework.
• Comparative assessment of visualizations derived from automatic topic extraction versus expert knowledge, including expert reviews and author discussions.

Main Results:

• The TextGenetics visualization effectively depicts the structure and chronology of text collections.
• The ConceptCircuit visualization successfully integrates crucial entities, enriching the understanding of the author's work.
• The visualization framework, coupled with advanced search, facilitates in-depth analysis and exploration of authorial texts.

Conclusions:

• The proposed visualization framework offers a powerful tool for scholars analyzing author-centric text collections.
• The developed techniques, TextGenetics and ConceptCircuit, provide unique insights into textual structure, chronology, and key entities.
• The study highlights the value of integrating visualization with sophisticated search capabilities for digital humanities research.