Welcome to the laboratory
Here is where I show off my nerdy side.
Explore
Credit Modelling
Objective: Create a machine learning model to allow a conservative investor to avoid people that will default on their loans.
Result: False positive rate of 9% achieved and a true positive rate of 24%.
Predicting House Sale Prices in Ames, Iowa
Objective: Use linear regression to predict the sale prices of houses in the city of Ames in Iowa, USA.
Result: RMSE of 28931.
Chinook Business Insights
Objective: Answer business questions and make the best decision for future business moves.
Result: The rock genre is popular in the USA. Czech Republic, United Kingdom, and India have great growth opportunities. Finally, album purchases account for 20% of total sales, so the offering should remain.
Hacker News Pipeline
Objective: Construct a data pipeline to retrieve the 100 most common keywords of the most popular articles on Hacker News.
Result: "Google", "Bitcoin", "programming", and "Python" make the headlines of the most popular articles.
Predict Bike Rentals
Objective: Identify the best machine learning model to predict the total number of bike rentals for a given time of the day.
Result: A random forest regressor with a minimum of 5 samples at a leaf node is the best model.
Fandango Movie Reviews
Objective: Determine if Fandango movie reviews are biased.
Result: Fandango is biased. The average rating is 4 out of 5, as opposed to 3 out of 5 from Metacritic.