ABSTRACT
The most common machine learning problems are the classification and regression. There exist more than 200 machine learning models for both the learning problems. The true performance of the model is the task performed with a statistical test by finding the confidence bounds. The machine learning model is used to predict the unseen patterns. Next, the predicted data is compared with the testing data. This comparison is drawn by various parameters. These evaluation parameters along with their statistics are also discussed. When the machine learning model results in minimum error using cross-validation, it seems to be the good model. The variables of the dataset are related to each other in one form or the other. It is very important to understand the relationship among the variables. There can be the number of correlation among the variables such as one variable completely depends upon on the other, or the two variables depend upon the third variable, and so on.
