https://github.com/aneeshmurali-n/nlp-emotion-classification-in-text

Develop machine learning models to classify emotions in text samples.
https://github.com/aneeshmurali-n/nlp-emotion-classification-in-text

bag-of-words data emotion-classification feature-extraction machine-learning naive-bayes natural-language-processing nlp nltk preprocessing python scikit-learn svm text-classification tf-idf tokenizer vectorizer

Last synced: 7 days ago
JSON representation

Develop machine learning models to classify emotions in text samples.

• Host: GitHub
• URL: https://github.com/aneeshmurali-n/nlp-emotion-classification-in-text
• Owner: aneeshmurali-n
• License: mit
• Created: 2024-10-11T18:30:11.000Z (26 days ago)
• Default Branch: main
• Last Pushed: 2024-10-16T18:02:38.000Z (21 days ago)
• Last Synced: 2024-10-31T11:06:34.101Z (7 days ago)
• Topics: bag-of-words, data, emotion-classification, feature-extraction, machine-learning, naive-bayes, natural-language-processing, nlp, nltk, preprocessing, python, scikit-learn, svm, text-classification, tf-idf, tokenizer, vectorizer
• Language: Jupyter Notebook
• Homepage:
• Size: 1.63 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# NLP Emotion Classification in Text

This project focuses on classifying emotions (fear, anger, joy) in text comments using Natural Language Processing (NLP) techniques and machine learning models.

## Project Overview

The project involves the following steps:

1. **Data Loading and Preprocessing:** Loads a dataset of comments and emotions, cleans the text by removing unnecessary characters and converting to lowercase, and tokenizes the text into individual words.

2. **Feature Extraction:** Creates numerical representations of the text using the Bag-of-Words (BoW) and TF-IDF methods.

3. **Model Development:** Trains two machine learning models - Naive Bayes and Support Vector Machine (SVM) - using the extracted features and emotion labels.

4. **Model Evaluation and Comparison:** Evaluates the performance of both models using metrics like accuracy, precision, recall, and F1-score, and compares their results.

## Results

The SVM classifier with BoW achieved the highest accuracy (91.7%) and demonstrated superior performance across various evaluation metrics. It is identified as the most suitable model for emotion classification in this project.

| Model | Accuracy | Precision (macro avg) | Recall (macro avg) | F1-score (macro avg) |

|---|---|---|---|---|

| SVM classifier with bow | 0.9166 | 0.92 | 0.92 | 0.92 |

| SVM classifier with TF-IDF | 0.9141 | 0.92 | 0.91 | 0.91 |

| Naive Bayes classifier with bow | 0.8905 | 0.89 | 0.89 | 0.89 |

| Naive Bayes classifier with TF-IDF | 0.8947 | 0.90 | 0.89 | 0.89 |

## Usage

To run this project:

Simply [Open Project in Google Colab](https://colab.research.google.com/github/aneeshmurali-n/NLP-Emotion-Classification-in-Text/blob/main/NLP_Emotion_Classification_in_Text.ipynb)

Otherwise:

1. Clone the repository.

2. Install the required libraries: `nltk`, `pandas`, `scikit-learn`.

3. Execute the Jupyter Notebook provided in the repository.

## Dataset

The dataset used in this project is located in this repository:  `nlp_dataset.csv`.

## License

This project is licensed under the MIT License.