This project focuses on sentiment analysis to assess mental health status based on textual data. Using advanced natural language processing (NLP) and machine learning, we aim to classify statements into seven categories: Normal, Depression, Suicidal, Anxiety, Stress, Bi-Polar, and Personality Disorder. This classification could aid mental health professionals in early identification and intervention.
Ensure you have the following packages:
pip install pandas numpy matplotlib seaborn nltk imbalanced-learn wordcloud scikit-learn xgboost- Text Cleaning: Tokenization, stemming, and removal of stop words.
- Vectorization: TF-IDF vectorization to convert text into numerical format.
- Oversampling: Addressed data imbalance using RandomOverSampler.
Algorithms: Tested multiple models, including:
- Logistic Regression
- Naive Bayes
- Decision Tree Classifier
- XGBoost Classifier
Evaluation: Models are evaluated using accuracy, precision, recall, F1-score, and confusion matrix.
Results: The best-performing model achieved an accuracy of approximately 81%. See the notebook for detailed metrics and model comparisons.
- Fine-tune models for improved accuracy.
- Explore additional NLP techniques for feature extraction.
- Expand to more mental health categories if more labeled data becomes available.