December 2018
Intermediate to advanced
318 pages
8h 28m
English
In this process, random selections are made from the class that has the majority of the data. This act is continued until both classes are balanced out. Though this method is good in terms of storage, but while random data reduction a lot of the important data points may get discarded. Another issue with this approach, is that it does not solve the problem of the dataset from which the random sample is picked being biased.
Read now
Unlock full access