Team members:
Aljaž Sebastjan Ahtik,63180041,[email protected]Jaša Kerec,63190138,[email protected]Luka Kuzman,63170014,[email protected]
Group public acronym/name: Avtoboti
Report (PDF) is available here
Install all the packages with pip.
cd src
pip install -r requirements.txt
Our datasets are placed inside folder /data. We have three different corpuses. English short stories (/data/english_short_stories), Slovenian short stories (/data/slovenian_short_stories), and Slovenian novels (/data/slovenian_novels). Inside /data folder we also have JSON files which contain named entities by stories, character frequencies by stories and resolved texts by stories.
The project is organized in a way, that you can always skip any step because all the results are stored in JSON files. This is great since some methods take really long time.
Open /src/sentiment_analysis_eng.ipynb and run all the cells. In the end, the results are stored in JSON files.
Open /src/sentiment_analysis_slo_coref.ipynb and run all the cells. In the end, the results are stored in JSON files.
cd src/sentex
python run.py <stories> <mode>
There are three possible stories values:
- ess: english short stories
- sss: slovenian short stories
- ssn: slovenian novels
There are three possible mode values (approaches):
- afinn
- vader
- bert
In the end again, the results are stored in JSON files.
cd src/cooccurrence
python run.py <stories> <mode>
The meaning of attributes is the same as in Sentiment Analysis.
cd src/cooccurrence
python run_visualize.py <story> <mode>
The mode is the same as in previous examples. And the story refers to a filename of the story file which you want to visualize (e.g. Henry_Red_Chief.txt).