Data Engineering for the Social World

Overview

Key facts

However, and perhaps surprisingly to many, the day-to-day of a data science project involves very little machine learning. It is often said, anecdotally, that data scientists spend 80% of their time cleaning and pre-processing data and only 20% building or deploying machine learning models. Therefore, whether you want to pursue a career in data science or to experience the data science way of doing things, it is crucial that you first learn how to handle data.

A person proficient at collecting, storing, and adequately pre-processing data is more likely to extract interesting insights from their data even before applying complex algorithms to a data set. This process is part of a data science and analytics subfield called data engineering.

London School of Economics and Political Science's Data Engineering for the Social World course will teach you to reason about data and how to collect real data from websites, APIs or other sources. It will also teach you the best practices for efficient data storage, the basics of SQL language, and the tools available in R to pre-process and reshape data. You will learn to put data in a "tidy" format, allowing you to re-purpose it for future analysis, be it for exploratory data analysis, visualisation or machine learning. You will also be free to choose the data sources that align the most with your interests.