Xente Fraud Detection Challenge
$4,500 USD
Accurately classify the fraudulent transactions from Xente's e-commerce platform
1201 data scientists enrolled, 547 on the leaderboard
Financial ServicesClassificationStructured
20 May 2019—23 September 2019
Data manipulation
published 21 Jun 2019, 09:05

Guys am having issues with preparing my training data for prediction , for-example the acountId , customerId etc how can i convert the into float values to allow the model understand well

Eric Peters you can use mean encoding, Weights of Evidence, frequency encoding etc . Search Online for various methods.

Mean/target encoding works quite well in most cases but presents a unique problem called target leakage. But not to worry, it has a minimal effect on performance.