Mental health disorders such as depression, anxiety, and suicidal tendencies have become major concerns in modern society. Social media platforms provide a large amount of textual data where individuals frequently express emotions, stress, and psychological conditions. This research presents a machine learning and Natural Language Processing (NLP)-based framework for analyzing social media posts to predict mental health conditions automatically. The proposed system performs text preprocessing using techniques such as lowercase conversion, stop-word removal, special character elimination, and text normalization. TF-IDF vectorization is applied to transform textual data into numerical feature vectors suitable for machine learning classification. Multiple machine learning algorithms including Logistic Regression, Naive Bayes, Support Vector Machine (SVM), Random Forest, and Dummy Classifier were implemented and compared using evaluation metrics such as Accuracy, Precision, Recall, and F1-score. The dataset consisted of approximately 40,000 social media records categorized into Suicide, Depression, Anxiety, and Normal classes along with demographic information such as age, gender, and location. Experimental results demonstrated that Random Forest and SVM achieved the best classification performance with an accuracy of approximately 57.6%. The study also performed demographic analysis to identify age-wise, gender-wise, and location-based mental health patterns. The proposed framework demonstrates the practical application of NLP and machine learning techniques for automated mental health prediction and awareness generation using social media analytics.

Keywords— Mental Health Analysis; Machine Learning; Natural Language Processing; TF-IDF; Social Media Analytics; Sentiment Classification