In this project I have done analysis on California Housing Data (1990)census. I have used different Machine Learning models like SVM, Random Forest Regression ,Multiple Regression, and XGBOOST . I also used deep learning and trained Artificial Neural Network with this dataset. The accuracy score shows that random forest regression is the winner and is best for prediction .The data is highly unbalanced as it lacks relatable features to the predicting feature.Thus Random Forest Model and the median income is the number one predictor of housing prices.The data set that i have ussed is avialable on kaggle. ANN model can be upgraded but it will comsume more time to train the model in epochs.Hence Random Forest Model is best fitting this dataset and is able to give better results than the other models I have trainned in this project.