First steps of cleansing

WenkyWong

Hi,

For a schoolproject I have to build a model that achieves the highest accuracy as possible. I have a retrieved a train set and test set from the professor, but I am wondering what would be the first steps of the data cleansing. 

I've attached the train_set and test_set as well. The label variable will be the price_category.

Anyone?

lionelderkrikor

Hi @WenkyWong,

As a starting point, I would use Turbo Prep : 

Click on CLEANSE --> AUTO CLEANSING

Turbo prep will clean automatically your dataset for you.
Then you can open the process RapidMiner has builded for you and see the different operations RapidMiner has performed on your initial dataset.

Then, I would use Auto-Model to train several models with your cleaned training dataset.

Hope this helps,

Regards,

Lionel