Logo Utrecht University

Translantis

BILAND data-mining tool

The BILAND data-mining tool

The Biland tool is best seen as a smart search engine. It is able to search on the base of either keywords of queries. These yield a number of hits, of which Biland is able to produce a quantitative analysis in a number of ways:

  • The number of hits tells something about the occurrence of certain words or combinations of words used in a certain period of time
  • A quantitative analysis of the sources used can give insight into in which social context to situate the occurrence of words
  • A word cloud generates the context within which certain words or combinations of words were used. The Biland tool is able to generate word clouds from singular articles, but also from the results of queries – in other words, from hundreds of articles together.
  • Within the word cloud and within articles sentiments can be highlighted. One can, in other words, see which words in the cloud had a negative connotation and which ones a positive
  • The Named Entity Recognition utility generates a word cloud consisting of words in the context of the keywords that are recognized as certain types of names: geographical names, names of persons or of institutions. Through the use of this tool we will be able to place the occurrence of certain ideas or debates within a geographical context, or to connect them to various persons or organizations.
  • Lastly, the temporal distribution visualization allows historians to discover patterns in documents’s publication dates. To enable quick recognition of atypical patterns, bursts within the histogram—time periods where significantly more documents were published compared to periods around that burst—are highlighted. Clicking on a burst yields a word cloud of that burst alone and a list of documents contained within that burst. This allows the historian to get an in-depth understanding of what each burst is about.

A complete manual of the BILAND tool can be found here:

WAHSP_manual_1.pdf (1.3 MB)