Skip to content

Latest commit

 

History

History
68 lines (41 loc) · 2.7 KB

README.md

File metadata and controls

68 lines (41 loc) · 2.7 KB

SciLander: Mapping the Scientific News Landscape

Repository Structure

  • Data: The data used in this study includes the computed triplets, the vocabulary files, and the source labels. The corpus is not provided directly in this repository but is openly accessible (see the Corpus section for more details).
  • Indicators: Code for individually computing the four indicators can be found in the folders: content_sharing_network, references, and semantic_shift.
  • Ensemble: Code for training unsupervised source embeddings using the indicators.
  • Experiments: All the experiments we conducted for our submission to ICWSM 2023.
  • Results: Plots of the experiments we conducted for our submission to ICWSM 2023.
  • Model: SciLander models as well as baselines models that we used in our experiments.

Setup

Set up the environment using Python VirtualEnv. From the root directory, run:

python -m venv venv/

Activate the environment just created:

source venv/bin/activate

Install the dependencies:

pip install -r requirements.txt

Corpus

The corpus used in this work was a combination of the NELA-GT-2020 and NELA-GT-2021 datasets, which can be downloaded in SQLite and JSON formats.

We filtered the corpus to retrieve only articles related to COVID-19 using a keyword matching procedure. The keywords can be found in data/CDC+COVID_vocab.txt. We selected all articles for which the title OR content had a match with at least one keyword from the list.

The SQLite database can be converted into CSV by using the script preprocessing/nela_to_csv.py.

Training a triplet loss model

Train a triplet loss using the triplets in data/triplets using the escript ensemble/train_source_embeddings.py.

cd ensemble
python train_source_embeddings.py

Computing triplets

Pre-computed triplets are provided in data/triplets. If you want to compute your own triplets, you can do so using the following scripts:

  • Jargon triplets: Use references/jargon_triplets.py.
  • Stance triplets: Use references/stance_triplets.py.
  • Content sharing triplets: Use content_sharing_network/csn_features.py.
  • Semantic shift triplets: Use semantic_shift/semantic_shift_triplets.py.

Pre-trained models

The pre-trained source embeddings models can be found in directory model. Look for any file with extension .emb.

Experiments

Most experiments require the pre-trained source embedding models from found in the model folder, in addition to the source labels found in the data folder.