Code for experiments conducted in the papers 'Who is hot and who is not? profiling celebs on twitter' and 'Fake or not: Distinguishing between bots, males and females' submitted to the Tenth International Conference of the CLEF Association (CLEF 2019) as part of the PAN author profiling task
Please cite these papers [bib] if you use this code.
Published results were produced in Python 3 programming environment on Linux Mint 18 Cinnamon operating system. Instructions for installation assume the usage of PyPI package manager.
To get the source code, clone the repository from github with 'git clone https://github.com/EMBEDDIA/PAN2019'
Data for the bot vs male vs female classification can be downloaded from here:
https://zenodo.org/record/3692340#.YAARLNYo-Uk
Data for the celebrity classification can be downloaded from here:
https://zenodo.org/record/3885373#.YAASeNYo-Uk
Install dependencies if needed: pip install -r requirements.txt
We have added a Jupyter notebook (see gender/src/example_usage.ipynb) in order to explain specific steps in the code.
To reproduce the results of celebrity classification published in the paper run the code in the command line using following commands:
Read data and generate features:
python parse_data.py --num_samples 100 --train_corpus pathToTrainCorpus --train_labels pathToTrainLabels --feature_folder pathToOutputFeatureFolder --all_data
Remove the '--all_data' flag if you want to reproduce the results on the evaluation set. If the flag is removed, 3837 examples are removed from the train set and used as a validation set.
Evaluate on development set:
python evaluate.py --feature_folder pathToOutputFeatureFolder
Generate test set predictions:
python test.py --input pathToTestCorpus --output pathToResultsFolder --feature_folder pathToOutputFeatureFolder
To reproduce the results of bot vs male vs female classification published in the paper run the code in the command line using following commands:
Read data and generate features:
python parse_data.py --train_corpus pathToTrainCorpus --feature_folder pathToOutputFeatureFolder
Evaluate on development set:
python evaluate.py --feature_folder pathToOutputFeatureFolder
Generate test set predictions:
python test.py --input pathToTestCorpus --output pathToResultsFolder --feature_folder pathToOutputFeatureFolder
Matej Martinc, Blaž Škrlj
- Knowledge Technologies Department, Jožef Stefan Institute, Ljubljana