Skip to content

An interactive Streamlit application for predicting the importance of academic abstracts using Microsoft Academic Graph

Notifications You must be signed in to change notification settings

anton164/predicting-citation-counts

Repository files navigation

aml-research-project

Please refer to the project report.

Set-up (Python 3.7+)

  1. Install dependencies
pip install -r requirements.txt
python -m spacy download en_core_web_sm
  1. Copy datasets to the repository manually
    • 250k.docs.jsonl (sample of 250k docs)
    • mag5.docs.jsonl (full dataset with 5 mill docs)

Run locally

streamlit run app.py

Or run individual streamlit pages:

  • Initial studies: streamlit run explore.py
  • Dataframe (feature) selection: streamlit run main.py
  • Experiment selection: streamlin run experiment_selection.py

About

An interactive Streamlit application for predicting the importance of academic abstracts using Microsoft Academic Graph

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages