Lexical Network
This visualization represents a map of individuals starting from their documents. It is based on the idea that authors can be described using their documents and that the vocabulary that authors share is an efficient way to draw a space. Individuals that make use of the same terms are situated nearby, beyond any social distance. The result is a continuo lexical space composed of individuals and terms.
The project is available at this URL https://rodighiero.github.io/LexicalNetwork/
To run the visualization on your local host
Clone the repository, which is built with JavaScript and Node. Then install the needed libraries by typing: npm install
The project can be tested on the local host by using npm run start and opening this URL in your browser. The project can also be built typing npm run build
Currently the project hosts two datasets, one of MIT thesis presenting the cartography of advisors, and one associated to the Digital Humanities Conference 2019. Both of them can be imported from the data folder typing:
node import-DH2019 or node import-MIT
The import sores in the same folder a file called docs.json that contains the documents in this way:
{
"id":
"title":
“body”:
“authors”: []]
}
The docs.json is then loaded and parsed using analysis.js, that collects texts by author and runs text analysis to compute the lexical distance. To run the analysis type:
node analysis
The analysis produced the authors.json, which is an array of objects composed in this form:
{
"id":
"docs":
[
"tokens": {
“token”: value,
}
},
Furthermore the analysis creates also the network used for rendering, which is composed of nodes.js and links.js