Курс от CourseraIn this course, you will be able to acquire data from diverse sources, perform thorough exploratory data analysis, assess and improve data quality, engineer meaningful features, and prepare clean, well-structured datasets for modeling and analysis. This course is designed to help you turn raw, imperfect data into reliable, analysis-ready inputs. You’ll learn how to systematically explore datasets, uncover patterns and anomalies, and apply appropriate techniques to clean, validate, and transform data. Through hands-on work with Python-based workflows, you’ll gain practical experience handling missing values, outliers, and inconsistencies, as well as generating statistical summaries and visual insights that guide decision-making. You’ll also develop the ability to engineer and select features, scale and encode variables, and create robust training, validation, and test splits that protect against data leakage. What makes this course unique is its end-to-end focus on data preparation as a critical foundation for successful analytics and machine learning. Drawing on expertise from multiple instructors, the curriculum blends exploratory analysis, data quality management, and feature engineering into a cohesive workflow. Whether you are new to data analysis or strengthening existing skills, this course equips you with repeatable, real-world techniques that improve model performance and analytical confidence across diverse use cases.
5 модулей · 79 учебных материалов

Преподаватель курса