I try to run SVM method based on previous regression model.
- First of all importing data movie_regression.csv and extracting into panda dataframes
- Missing value treatment by using mean value
- Dummy variable treatment by replacing string value with binary variable
- Define X and y variable
- train the data using test train split syntax. We only take 20% of available data as testing data
- Standardizing data by converting mean and variance of each data into 0 and 1
- Predict y value using svm model
- Find out the model performance accuracy. R2 for y train prediction and y test prediction stood above 0,5 which indicates a good model