NUS Group 252

Inspiration

What it does

How we built it

Challenges we ran into

-large number of columns -messy data, many NaN values -imbalanced target data

Accomplishments that we're proud of

• managed to make sense of the features using domain knowledge to feature engineer new columns
• managed to reduce the dimension of the dataset, keeping the important features managed the imbalanced target variables using SMOTE oversampling

What we learned

What's next for NUS Group 252

Built With

• matplotlib
• numpy
• pandas
• python
• scikit-learn
• tensorflow