This repository contains the source code for the web application driving the JSTOR Labs Topicgraph demonstration app. The Topicgraph app was developed as part of the "Reimagining the Monograph" project conducted by JSTOR Labs during the Fall of 2016.
This application was developed using the Angular2 javascript framework and can be installed and run using either the Node Package Manager (NPM) or yarn.
You can use npm, but it's recommended to use yarn as it installs a lot faster and has other benefits https://yarnpkg.com/ . Make sure you are using yarn version 0.16.0 or newer (check with 'yarn --version')
git clone [email protected]:JSTOR-Labs/topicgraph.git
cd topicgraph
yarn
yarn startThis application extends the angular-webpack2-starter seed project, which provides support for Angular2, Ahead of Time (AOT) compilation, server-side rendering (Angular Universal) and many other features.
For this project the Owl Carousel, Font Awesome icons, pdf.js and D3 libraries were added.
REST APIs provided by JSTOR Labs are used for in the analysis and topic inference of the monographs.
These APIs should be considered experimental and subject to change.
select - Retrieves monograph metadata and serialized thumbnail images from back-end service. If a
document ID is provided only the metadata for that single document is returned, otherwise the
metadata and thumbnails for all public monographs is returned.
getVisualizationData - Retrieves topic data for a monograph for use in visualization. The topic data
is generally available at multiple levels of granularity. If unspecified the level defaults to level
'0', or the most coarse.
getWordCoords - Retrieves the full text for a monograph with corresponding woord coordinates.
This word coordinate text is used in word highlighting.
getTopicWords - Retrieves the words associated with a specified topic.
submit - Submit a PDF for off-line analysis. An email notification is generated when the
document processing is complete.