QuaXP - Data Quality Explored
The project aims to raise or sharpen public awareness on data quality problems in the context of machine learning.
Go to courseWhat to expect
Did you know that most of the hiring processes in big companies are complemented by machine learning algorithms? That some banks use machine learning to predict whether or not you will be able to reimburse a credit? That most online customer support usually start the interactions with an intelligent agent rather than a human? Because of the vast amount of applications we are confronted with in our daily life, a basic knowledge of data science is essential to understand how decisions are made – QuaXP helps you with that.Learning objectives
The learners understand how to measure and assess errors in a dataset, such as noise, missing or corrupted data. They learn techniques for data preparation and data cleaning, to increase the quality of a dataset for its further use in a machine learning experiment.
Learners with a basic knowledge of programming use code and learn methods from diverse libraries of Python. Learners without prerequisite can directly visualize the results of the data and their changes on interactive graphs.
Learning method
QuaXP is comprised of 3 chapters, each exploring data quality for a type of data: the first chapter deals with numerical data, the second with image data, and in the third chapter, we explore text data. An introduction contains introductory content that helps understanding the concepts of the course material. Each chapter contains subsections with a video, and at the end of each page of the book, a quiz is implemented to test the newly acquired knowledge. At the end of each chapter, a practical task is available for the learners of the advanced content. A general review quiz is here for everyone.
More info
Anna Lainé