IronyDetectionInTwitter

URL: https://github.com/NIHRIO/IronyDetectionInTwitter

Description: A Simple and Accurate Neural Network Model for Irony Detection in Twitter

Methods

This model utilizes a simple neural network architecture of a Multilayer Perceptron (MLP) with various input features, including:

• Lexical features
• Syntactic features
• Semantic features
• Polarity features

The model was developed for the SemEval 2018 Task 3: Irony Detection in English Tweets. It was evaluated using two subtasks:

• Subtask A: Binary irony classification (ironic vs. non-ironic).
• Subtask B: Multi-class irony classification.

Results

• The model achieved high performance in both binary and multi-class irony detection subtasks.
• It ranked at least third based on the accuracy metric and fifth using the F1 metric.

Dataset

• The dataset used is from SemEval 2018 Task 3 on Irony Detection in English Tweets.
• Specific details about the dataset (e.g., number of samples, distribution of classes) are not provided in the repository.