Data quality has always been a pervasive problem in the industry. The presence of incorrect or inconsistent data can produce misleading results of your analysis. Implementing better algorithm or building better models will not help much if the data is not cleansed and prepared well, as per the requirement. There is an industry jargon called data engineering that refers to data sourcing and preparation. This is typically done by data scientists and in a few organizations, there is a dedicated team for this purpose. However, while preparing data, a scientific perspective is often needed to do it right. As an example, you may not just do mean substitution to treat missing values and look into data distribution to find more appropriate values to substitute. Another such example is that you may not just look at a box plot or scatter plot to look for outliers, as there could be multivariate outliers which are not visible if you plot a single variable. There are different approaches...