What does data cleaning refer to?

Data cleaning refers to the process of addressing inaccuracies and problems within a dataset to ensure that it is accurate, consistent, and usable for analysis. This process involves identifying and correcting errors, handling missing values, removing duplicates, and standardizing data formats. The goal is to enhance the quality of the data, which is fundamental for any data analysis or modeling tasks. Clean data leads to more reliable insights and better-informed decision-making.

Analyzing data patterns, encoding categorical data, and visualizing data are all important steps in the data analysis workflow, but they do not directly pertain to the cleaning process itself. They come into play after the data has been cleaned, to be used effectively for analysis, modeling, or presentation.