A journal recommender tool built on the Directory of Open Access Journals
This application suggests open access journals based on their similarity to a draft abstract submitted by the user. It is meant for authors who are trying to discover suitable target journals for their work. The results are meant to be serendipitous; the goal is to uncover unexpected but relevant journals.
The application is built with Flask, combined with "serverless" infrastructure for data analysis. The Flask application calls a Google Cloud Function asynchronously. Most of the computationally intensive work is done by the Cloud Function. Specificaly, the Cloud Function does similarity calculations using spaCy and returns a similarity score for each potential target journal.
Presented at the 18th Annual CUNY IT Conference. New York, NY. December 5, 2019.
This project is partly supported by a grant of Google Cloud Platform credits.