We often come across datasets where not all of the values are available for specific variables/attributes. This can happen for several reasons, such as ignored questions in a survey, typing errors, a device malfunctioning, and so on. Encountering these missing values is expected in a data mining project, and dealing with these values is essential.
Missing value imputation occupies most of a data scientist's time. There are various ways through which we can impute missing values. The deciding factor is what to use when attributing these kind of unavailable values. The process on deciding what and when to use for imputing missing values is a talent and comes from experience in working with data. Sometimes it is better to remove ...
