MVP:

Plot heatmap of correaltion for better visualization:

Plotting the correlation matrix of the dataset

Attack Category:

The name of each attack category. In this data set, there are total nine categories of attack and normal is non-attack.
In train data most occured attack data categories are "Generic", "Exploits", Fuzzers", "DoS" and "Reconnaissance"

Baseline Modeling Result:

LogisticRegression:

Accuracy:

Traning Score: 0.7865
Validation Score: 0.7855

Precision:

Traning Score: 0.8395
Validation Score: 0.8386

Recall:

Traning Score: 0.9669
Validation Score: 0.9679

Confusion Matrix