In short
Data cleansing is finding and fixing problems in data before analysis, such as duplicates, missing values, inconsistent formats, invalid entries and outliers.
Key takeaways
- It is part of the transform step in extracting, transforming and loading data.
- Changes should be documented so the analysis can be reproduced.
- Outliers should be investigated, not just deleted, because they may be the most important items.
Where it's tested
BAR: Data Analytics for Accounting — study notes, a five-minute revision sheet and practice questions.
Practise: 3 questions
Question 1 of 3 · 0 answered
An analyst builds a model that recommends the optimal inventory reorder quantity for each SKU based on demand forecasts and carrying costs. What type of analytics is this?