I started my career as a MOLAP cube builder which left space
for my creativity to improve the model that was in the data warehouse. It was
always a great feeling to create a tool that could give the users ability to
create reports on their own.
Data Explorer (CDE) is the FBI’s Uniform Crime Reporting (UCR) Program’s dynamic solution to presenting crime data in a more
immediate venue that reflects the constant change in the nation’s crime
pages provide a view of estimated national and state data, reported
agency-level crime statistics, and graphs of specific variables from the
National Incident-Based Reporting System (NIBRS).
It was a very nice trip down memory lane to hear a talk about OLAP cubes. I spent a great amount of my professional career building MOLAP cubes and this talk made we want to build one again just for fun.
The following tutorial is going to describe my Power BI report that is using the new Key Influencers visual to enter in a machine learning competition on Kaggle using only Power BI.
The sinking of the RMS Titanic is one of the most infamous
shipwrecks in history. On April 15, 1912, during her maiden voyage, the
Titanic sank after colliding with an iceberg, killing 1502 out of 2224
passengers and crew.
Solving this task at first I started with the standard Decision Tree, without any tuning. Then I get into GridSearchCV and RandomizedSearchCV for the best parameters. But after tweaking the model with these validations, I still couldn’t get higher than 79%. RandomForest didn’t help either.
That’s when I found XGBoost, a powerful model, getting more and more attention in machine learning. With it, I could go over 80%.