This paper focuses on developing a machine learning model for the classification of brain tumors into three categories: No tumor, Benign Tumor (Non-cancerous) and Aggressive tumor. Early and accurate classification of brain tumors is important for deciding the right treatment and improving patient outcomes. We used a dataset containing patient details like age, gender, blood pressure, glucose level, cholesterol, family history, and symptoms such as seizures, nausea, and vision problems. A Multi-layer Perceptron was trained on this data, and Synthetic Minority Oversampling Technique (SMOTE) was applied to solve the issue of class imbalance. The model was further improved using GridSearchCV for hyperparameter tuning. We then deployed the model using Streamlit, creating an interactive web application that allows both single-patient prediction through sliders and dropdowns, as well as bulk predictions through CSV upload. The system is designed to assist healthcare professionals in making quick and accurate decisions and to provide a simple interface for users without a technical background. The prediction of the Tumor along with its type for Single as well as Bulk patients input using MLP with almost 80% accuracy.