{"id":11984,"date":"2026-08-10T10:39:38","date_gmt":"2026-08-10T10:39:38","guid":{"rendered":"https:\/\/nokobox.com\/index.php\/item\/oreilly-cleaning-data-for-effective-data-science-2025-6\/"},"modified":"2026-08-10T10:39:38","modified_gmt":"2026-08-10T10:39:38","slug":"oreilly-cleaning-data-for-effective-data-science-2025-6","status":"publish","type":"digital_item","link":"https:\/\/nokobox.com\/index.php\/item\/oreilly-cleaning-data-for-effective-data-science-2025-6\/","title":{"rendered":"Oreilly \u2013 Cleaning Data for Effective Data Science 2025-6"},"content":{"rendered":"
\n

Description<\/span><\/h2>\n

Cleaning Data for Effective Data Science. The first and most critical step in the data science process, creating useful data, is taught. This step is often the most time-consuming part of the job. In this course, participants will learn how to differentiate between different data formats, work with tabular and hierarchical data, extract data from reconstructed sources, and detect anomalies. They will also learn how to assess data quality, correct missing or problematic data, identify and manage outliers, and replace unreliable data with acceptable values. Finally, the principles of sampling are covered. The prerequisite for the course is proficiency in a programming language used in data processing and machine learning. Course topics include ingesting data from tabular (e.g., CSV, spreadsheets) and hierarchical (e.g., XML, JSON, NoSQL databases) formats, as well as extracting data from sources such as web pages and PDF documents. Other courses cover anomaly detection, systematic assessment of data quality by considering biases and normalization techniques, and finally, value substitution using methods such as normalization, trend reflection, and sampling techniques. This course provides the tools necessary to create a comprehensive dataset ready for the next steps in the data science pipeline.<\/span><\/p>\n

What you will learn<\/span><\/h3>\n