This Saturday, in the fourth webinar we are discussing how analyst and data scientist use #Python for predictive analytics. Register for future webianrs here: https://digitalmarketinginstitute.in/webinars/
In this #live and interesting session learn how to use Python with #JupyterNotebook to find insights that aid organizations take critical decisions. This is a hands-on session with cool demos.
#Analytics #DataScience #DataAnalytics #BusinessAnalytics #webinars #bigdata
- Python
- Jupyter notebook
- Linear Regression
- Multi-regression
- Question & Answer
5. Building a Model
- Data collection
- Data Cleaning(cleaning, scaling,normalization etc.)
- Split the data in train and test
- Choose a model
- Training the model(Fit to the model)
- Test your model
- Evaluate your model
6. Building a Linear Regression Model
1. x=df[[ 'Avg. Session Length', 'Time on App', 'Time on Website', 'Length of
Membership']]
2. y=df['Yearly Amount Spent']
3. x_train,x_test,y_train,y_test=train_test_split(x,y,test_size=0.4,random_state=101)
4. lm=LinearRegression()
5. lm.fit(x_train,y_train)
7. Things to Remember - Linear Regression
- Multicollinearity in multiple linear regression
- Normality in the distribution
- Independency in values of output variables
8. Linear Model evaluation parameters
- R-squared/Adjusted R-square
- Measure the percentage of variation in the dependent variable explained by the predictor
- Root Mean Square Error(RMSE)
- Measure the sq root of average of the squared difference between the observed and
actual value. Lower the better
- Mean Absolute Error
- Measure the average of absolute difference between the observed and the actual value