Description
This paper describes the use of two open source software libraries for extracting meaning from a story–like textual narrative and displaying it in an intuitive and interactive manner to users via a set of word cloud visualizations. The first library is Stanford’s CoreNLP, a Java API that provides natural language processing services such as sentence recognition, tokenizing, parts-of-speech identification, dependency parsing, named entity recognition, and coreference resolution. The second is D3, a JavaScript API built on scalable vector graphics which provides powerful data visualization capabilities. The paper presents an application that uses CoreNLP to extract key story elements from a body of text (subjects, actions, objects, places, times, and other contextual features), and uses D3 to display these elements and their interactions in an integrated and interactive set of word clouds.
Recommended Citation
Mitri, Michel, "Story Visualization via NLP and Word Clouds" (2017). AMCIS 2017 Proceedings. 8.
https://aisel.aisnet.org/amcis2017/SemanticsIS/Presentations/8
Story Visualization via NLP and Word Clouds
This paper describes the use of two open source software libraries for extracting meaning from a story–like textual narrative and displaying it in an intuitive and interactive manner to users via a set of word cloud visualizations. The first library is Stanford’s CoreNLP, a Java API that provides natural language processing services such as sentence recognition, tokenizing, parts-of-speech identification, dependency parsing, named entity recognition, and coreference resolution. The second is D3, a JavaScript API built on scalable vector graphics which provides powerful data visualization capabilities. The paper presents an application that uses CoreNLP to extract key story elements from a body of text (subjects, actions, objects, places, times, and other contextual features), and uses D3 to display these elements and their interactions in an integrated and interactive set of word clouds.