Scikit-Learn Essentials: Data Preprocessing Techniques

#DataModelingWorkshops #AIandMLPreparation #IEEEModelthon1_0 #KaggleCompetition #DataPreprocessing #DataCleaning #MLAlgorithmOptimization #DataPipelineCreation #CategoricalDataEncoding #DataImputation #OutlierRemoval #MissingValuesHandling #DataReadiness #DataPreparationPhase #MLModelTraining #SupervisedLearning #DataScienceTechniques #DataCleaningWorkshop #MachineLearningReadiness #DataDrivenSolutions

Workshop Series Introduction

This workshop is one of a series of workshops that aim to enhance understanding of data modeling processes for those interested in artificial intelligence and machine learning. Specifically, it is designed to prepare participants for the first edition of the upcoming IEEEModelthon 1.0 on Kaggle.

Workshop Overview

The next phase after applying data preprocessing is to run the machine learning algorithm. The result of this phase directly affects the performance of the ML algorithm. So, it is critical to have the data ready in its final form with no wrong data, no missing values, no outliers, no duplicated rows, etc.

Purpose

The primary purpose of this phase is to have the data in its final form, ready to use for the ML algorithm. Some ML algorithms work well with categorical data and apply embedded encoding techniques and others do not (we should manually encode categorical columns), some of them deal well with missing values and apply an embedded imputation and some do not, and very many considerations!

Objective and Outcome