Recently, social media has taken over the majority of the day. This is mostly because everyone expresses emotions or responds to tweets or posts with comments. As the individual expresses their thoughts, a large amount of data can be generated. This data is utilized to analyze people's sentiments and emotions and predict their mental health. The fundamental aim of this research is to create a prediction model that utilizes available data to predict an individual's mental health. The data set utilized includes Twitter data. To convert the text-based data into numerical values, Term Frequency-Inverse Document Frequency (TF-IDF) is utilized. Then, MNon-negative Matrix Factorization (NMF) and Clustering are performed in these numerical values to decompose large data into non-negative matrices and group similar data respectively. This study has significant implications for mental health professionals, as it offers a scalable and non-invasive tool for early detection and intervention in mental health issues. Different machine learning algorithms such as Random Forest (RF), Decision Tree (DT), Support Vector Classifier (SVC), Multinominal Naive Bayes (MNB), and XGBoost have been used. The result of the study shows promising predictive capabilities for identifying mental health conditions based on tweets. Among these, RF and DT achieved the highest accuracy of 97.4% in classifying individuals with mental health challenges. In conclusion, this study presents a novel approach to predicting mental health outcomes using Twitter data.