Google Colab Notebook for Insurance Data
Budget: $10 – $30 USD
Perform the following tasks in a Google colab notebook:
Data Exploration
-Load the insurance data.
-Print out the names of its columns.
-Print out the head of the dataset.
-How many null values are present?
-Use the pandas ‘describe’ command to print out summary information about the dataset.
-What are the correlations between the columns of the dataset? (You may want to encode the categorical features at this point, which can be done using the LabelEncoder from scikit-learn.)
-Produce a heatmap to display these.
-In a few sentences, provide an interpretation of the heatmap.
Data Modeling – Predicting Insurance Charges
-Produce an 80-20 training-testing split of your data.
-Train a Linear Regression model to predict ‘charges’ using the training set.
-Print out 5 ‘charges’ of the training data along with your model’s predicted ‘charges’ on these.
-Compute the MSE score of your model on the test set.
-In a few sentences, explain the MSE score.
-Compute the R2 (coefficient of determination) scores of your model on the test set.
-In a paragraph, explain the R2 score and then interpret the R2 score you have computed.
Communicating and Visualizing the Results
-Produce a plot to showcase the effectiveness of your model.
-Present a summary of your findings and their implications in a concluding paragraph.-
Data Exploration
-Load the insurance data.
-Print out the names of its columns.
-Print out the head of the dataset.
-How many null values are present?
-Use the pandas ‘describe’ command to print out summary information about the dataset.
-What are the correlations between the columns of the dataset? (You may want to encode the categorical features at this point, which can be done using the LabelEncoder from scikit-learn.)
-Produce a heatmap to display these.
-In a few sentences, provide an interpretation of the heatmap.
Data Modeling – Predicting Insurance Charges
-Produce an 80-20 training-testing split of your data.
-Train a Linear Regression model to predict ‘charges’ using the training set.
-Print out 5 ‘charges’ of the training data along with your model’s predicted ‘charges’ on these.
-Compute the MSE score of your model on the test set.
-In a few sentences, explain the MSE score.
-Compute the R2 (coefficient of determination) scores of your model on the test set.
-In a paragraph, explain the R2 score and then interpret the R2 score you have computed.
Communicating and Visualizing the Results
-Produce a plot to showcase the effectiveness of your model.
-Present a summary of your findings and their implications in a concluding paragraph.-
Related categories:
Python
Statistics
R Programming Language
Statistical Analysis
SPSS Statistics
NumPy
Data Analysis
Pandas