Summary
Linked Jazz is a research project which provides data visualization for closeness of relationships among many well-known and lesser-known jazz artists. The visualization includes an interactive network map, with each artist representing a node, and a page where the user decides to rate the closeness of relationship based on text sampled from interviews with the various artists.
Goals
- To uncover meaningful connections between documents and data related to the personal and professional lives of jazz artists.
- To develop broadly applicable tools and methods for working with Linked Open Data.
Sources/Assets
The developers of Linked Jazz have collected a variety of data sources for their visualization, including:
- Transcripts of oral interviews (obtained from various college jazz archives)
- DBPedia (Wikipedia information formatted according to Linked Open Data)
The oral interviews and transcripts are sourced from a wide spectrum of institutions, such as Hamilton College, Rutgers, the Smithsonian, UCLA, and more. The DBPedia is an open source database which anyone can draw from if they know how to send the appropriate URL requests. I came up with a question while researching DBPedia: What API would be the most efficient when working with DBPedia? What would be the most efficient function for retrieving data from DBPedia for a given API?
Processes
Linked Jazz developed and used many different tools in the process of creating this project, including a transcript analyzer, which identifies names in the interview transcripts, a Curator tool to create a directory of personal names, and an API wrapper used for matching entities called Ecco. The wrapper reorganizes the retrieved data and ranks it by confidence of matches (accuracy).
Presentation
The website contains two different tabs for the presentation and visualization of their collected data. One is an interactive network map which shows the overall network of individuals, and shows the overall network of a certain individual if the user clicks on them. The other is a user input-focused page where the user clicks on a certain individual and rates their closeness to the other jazz musicians based on quotes from interview transcripts.
Related Academic Fields
This project could prove useful when constructing biographies of famous or otherwise significant individuals. It excels at visually communicating the degree of closeness that two figures expressed. Linked Jazz also has a comprehensive API that expands its availability from jazz artists to any group of figures in history. This program and the associated tools could lend their accessibility to the social sciences.
