Demonstrated on the Medical Officer of Health reports at the Wellcome Library
Vane, O., 2018, May. Text Visualisation Tool for Exploring Digitised Historical Documents. In Proceedings of the 2018 ACM Conference Companion Publication on Designing Interactive Systems (pp. 153-158). ACM.
Timeline design for visualising cultural heritage data, Chapter 4.
Built in JavaScript / D3.js with ElasticSearch
Data visualisation is often thought of as a useful technique to look at patterns in quantitative data: but what about the qualitative? Steptext provides a helpful structure for surveying texts by time, revealing and emphasising what is being said. The visualisation tool enables a historian to quickly survey a document collection around themes they are interested in while staying close to the original texts. Simple visuals help promote transparency.
The interface works as follows: a user searches for a keyword they are interested in. The interface then visualises instances of that keyword across all the documents, mapped by time horizontally. This way, the user can easily trace commentary through time.
Steptext is demonstrated on The Medical Officer of Health reports: a set of 5,500 19th Century public health reports digitised by the Wellcome Library. These reports mostly feature narrative text and have been converted to digital text files.
To read more about Steptext including evaluations with historians, read the short paper Text Visualisation Tool for Exploring Digitised Historical Documents, or my longer thesis chapter Steptext: Medical Officer of Health reports.