https://github.com/sayamalt/mental-health-classification-using-fine-tuned-distilbert

Successfully established a multiclass text classification model by fine-tuning pretrained DistilBERT transformer model to classify several distinct types of mental health statuses such as anxiety, stress, personality disorder, etc. with an accuracy of 77%.
https://github.com/sayamalt/mental-health-classification-using-fine-tuned-distilbert

data-visualization deep-learning distilbert-fine-tuning distilbert-model model-evaluation model-inference model-training-and-evaluation multiclass-text-classification natural-language-processing text-classification text-preprocessing text-tokenization

Last synced: 4 months ago
JSON representation

Successfully established a multiclass text classification model by fine-tuning pretrained DistilBERT transformer model to classify several distinct types of mental health statuses such as anxiety, stress, personality disorder, etc. with an accuracy of 77%.

• Host: GitHub
• URL: https://github.com/sayamalt/mental-health-classification-using-fine-tuned-distilbert
• Owner: SayamAlt
• License: apache-2.0
• Created: 2025-01-06T16:20:23.000Z (5 months ago)
• Default Branch: main
• Last Pushed: 2025-01-06T16:52:04.000Z (5 months ago)
• Last Synced: 2025-01-06T17:49:39.385Z (5 months ago)
• Topics: data-visualization, deep-learning, distilbert-fine-tuning, distilbert-model, model-evaluation, model-inference, model-training-and-evaluation, multiclass-text-classification, natural-language-processing, text-classification, text-preprocessing, text-tokenization
• Language: Jupyter Notebook
• Homepage:
• Size: 2.07 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Sentiment Analysis for Mental Health

## About the Project

This project utilizes a comprehensive dataset curated to analyze and predict mental health statuses from various textual statements. The dataset, compiled from multiple sources, serves as a robust resource for tasks such as chatbot development and sentiment analysis.

## Dataset Overview

The dataset integrates raw data from the following Kaggle sources:

- [3k Conversations Dataset for Chatbot](https://www.kaggle.com/datasets)

- [Depression Reddit Cleaned](https://www.kaggle.com/datasets)

- [Human Stress Prediction](https://www.kaggle.com/datasets)

- [Predicting Anxiety in Mental Health Data](https://www.kaggle.com/datasets)

- [Mental Health Dataset Bipolar](https://www.kaggle.com/datasets)

- [Reddit Mental Health Data](https://www.kaggle.com/datasets)

- [Students Anxiety and Depression Dataset](https://www.kaggle.com/datasets)

- [Suicidal Mental Health Dataset](https://www.kaggle.com/datasets)

- [Suicidal Tweet Detection Dataset](https://www.kaggle.com/datasets)

### Data Overview

The dataset contains statements tagged with one of the following seven mental health statuses:

- **Normal**

- **Depression**

- **Suicidal**

- **Anxiety**

- **Stress**

- **Bi-Polar**

- **Personality Disorder**

### Data Features

- **`unique_id`**: A unique identifier for each entry.

- **`Statement`**: The textual data or post.

- **`Mental Health Status`**: The tagged mental health status of the statement.

### Data Collection

The data is sourced from various platforms, including:

- Social media posts

- Reddit posts

- Twitter posts

Each entry is carefully tagged with a mental health status, making it a valuable asset for:

- Developing intelligent mental health chatbots.

- Conducting detailed sentiment analysis.

- Researching mental health trends and patterns.

## Usage

This dataset can be used for:

- **Chatbot Development**: Build chatbots focused on mental health support.

- **Sentiment Analysis**: Understand and predict mental health conditions based on textual data.

- **Academic Research**: Study patterns and trends in mental health.

### Example Applications

1. Training machine learning models for mental health sentiment prediction.

2. Developing tools to monitor and provide mental health support in real time.

3. Research studies focused on understanding mental health trends across demographics.

## Acknowledgements

This dataset was created by aggregating and cleaning data from various publicly available Kaggle datasets. Special thanks to the original dataset creators for their invaluable contributions.

---

For any inquiries or contributions, please feel free to open an issue or submit a pull request.