如何提高入住率预测的准确性？

Question

I have a project that's aimed to predict the amount of occupants at my local gym given the date and weather.

Here's my Kaggle kernel

I have two datasets, occupants on a given hour and weather on a given hour. My process is that I combine these two datasets, and using Occupants as the target. However, when I implement a regression algorithm I can only reach a prediction score of 57%.

I'd love any advice on how to modify my solution to achieve better predictions?

Thank you.

Answer 1

To Improve the accuracy:

• Do more feature engineering.
• You can convert your categorical variable into one hot encoded values.
• You can make one more feature stating morning, afternoon, evening, night using hour from timestamp and also you can add 'weekday/weekend' column.
• You have not done EDA much. Check for outliers and remove them.