Détection et recodage d'éléments paralinguistiques

Projet de master en analyse textuelle avec Python, entrepris à l'UNIL sous la direction d'Aris Xanthos, SA2023.

1. Purpose

This programme is designed for detecting and recoding paralinguistic elements in chat data. Please read final-report.pdf for more detail about the work that I have undertaken.

2. Installation

Install the following required dependencies:

• pip install spacy
• pip install emoji
• pip install nltk
• pip install tqdm

Download the following resources by uncommenting the following code:

nltk.download('punkt') nltk.download('averaged_perceptron_tagger') nltk.download('universal_tagset')

3. Initiation

1. Ensure you have prepared the text file you wish to encode.
2. Run the script from your terminal: python main.py.
3. Input the file name, format type and if you wish to include an external file of reserve words as prompted.
4. A progress bar will show you the status of the encoding process.
5. Once the programme has run, you will be asked whether you wish to save the encoded file externally.

4. Package

To use the programme functions externally and independently in your own programme, first import the script into your programme with import main.py and then use the functions as required.