Dutch Sarcasm Detector

🚀 Dutch Sarcasm Detector

The Dutch Sarcasm Detector is a text classification model designed to detect sarcasm in news article titles. It is fine - tuned on [GroNLP/bert - base - dutch - cased](https://huggingface.co/GroNLP/bert - base - dutch - cased). The training data includes a ready - made dataset from Kaggle and scraped data from the Dutch sarcastic newspaper (De Speld).

Labels: 0 -> Not Sarcastic; 1 -> Sarcastic

✨ Features

• Built on a pre - trained Dutch BERT model for effective sarcasm detection.
• Utilizes a combination of public datasets and scraped data for training.

📦 Installation

No specific installation steps are provided in the original README. If you want to use the model, you can follow the code example below to load the model.

💻 Usage Examples

Basic Usage

from transformers import AutoModelForSequenceClassification
from transformers import AutoTokenizer
import string

def preprocess_data(text: str) -> str:
   return text.lower().translate(str.maketrans("", "", string.punctuation)).strip()

MODEL_PATH = "helinivan/dutch-sarcasm-detector"

tokenizer = AutoTokenizer.from_pretrained(MODEL_PATH)
model = AutoModelForSequenceClassification.from_pretrained(MODEL_PATH)

text = "We deden een man een nacht in een vat met cola en nu is hij dood"
tokenized_text = tokenizer([preprocess_data(text)], padding=True, truncation=True, max_length=256, return_tensors="pt")
output = model(**tokenized_text)
probs = output.logits.softmax(dim=-1).tolist()[0]
confidence = max(probs)
prediction = probs.index(confidence)
results = {"is_sarcastic": prediction, "confidence": confidence}

Output:

{'is_sarcastic': 1, 'confidence': 0.8915400505065918}

📚 Documentation

Source Data

Datasets:

• Dutch non - sarcastic data: [Kaggle: Dutch News Articles](https://www.kaggle.com/datasets/maxscheijen/dutch - news - articles)

Scraped data:

• Dutch sarcastic news from De Speld

Training Dataset

• helinivan/sarcasm_headlines_multilingual

Codebase

• Git Repo: [Official repository](https://github.com/helinivan/multilingual - sarcasm - detector)

📄 Performance