Confusion Matrix and it’s implementation in Cyber Crime Cases
Cyber attacks have become one of the biggest problems in the world. They wreak havoc on countries and individuals on a daily basis. Increasing cyber attacks also lead to cyber crime. Key elements in the fight against crime and criminals are to identify cyber criminals and to understand the methods of attack. Finding and avoiding cyber attacks is a daunting task. However, researchers have recently been able to solve these problems by building security models and making predictions on artificial intelligence.
There is one of the metrics used to classify these cyber crime events named as Confusion Matrix. So before knowing correlation between Confusion Matrix and Cyber Crime, let’s deep dive to know about Confusion Matrix.
What is Confusion Matrix?
The confusion matrix is a summary of the predictive effects on the separation problem. The number of positive and negative predictions is summarized by the calculation values and divided by each category. This is the key to the confusion matrix.
A Confusion matrix is matrix of 2 X 2 which consist of True positive, True Negative, False Positive and False Negative.
📌True Positive-: True positive (TP) refers to the actual value and the predicted value is 1 means True.
For example- You predicted that a man is deaf and he actually is.
📌True Negative-: True negative (TN) refers to the actual value and the predicted value is 0 means False.
For example- You predicted that a man is not deaf and he actually is not.
📌False Positive-: False positive (FP) refers that the actual value is 0 but the predicted value is 1. False Positive is a Type 1 error. It is most dangerous error.
For example- You predicted that a man is deaf but he actually is not.
📌False Negative-: False Negative (FN) refers that the actual value is 1 but the predicted value is 0. False negative is a Type 2 error.
For example- You predicted that a man is not deaf but he actually is.
What is Accuracy?
Accuracy represents the number of correctly classified data instances over the total number of data instances.
Formula to Calculate Accuracy-:
What is Precision?
It is the average of the best predicted samples in the total number of positively predicted samples.
Formula to Calculate Precision-:
What is Recall?
Recall literally is how many of the true positives were recalled (found), i.e. how many of the correct hits were also found. Recall is also known as sensitivity or true positive rate.
Formula to Calculate Recall-:
What is F1 Score?
F1 score helps to measure Recall and Precision at the same time. It is a Harmonic Mean of precision and recall and is a better measure than accuracy.
Formula to Calculate F1 Score-:
Cyber Crime and Confusion Matrix
Let’s suppose a scenario in an Organization Theft of financial or card payment data cyber attack is going to happen. The Organization members proposed a model regarding this event to monitor this cyber attack.
The confusion matrix of proposed model where the training size was 0.8 and the test size was 0.2. With this, we know how many cases have been properly divided and how many have been judged incorrectly. It means that we can find the true and wrong benefits of illegal and false models separated by the model. When the test size was 0.40 it predicts three cases of copyright theft. Now, As soon as the test size is increased to 0.60, It predicts that seven identity theft cases have been labeled as infringement and one copyright infringement has been incorrectly terminated as data theft and one identity theft case filed as hacking. The total number of cases divided incorrectly when the test size was increased to 0.8. It shows that fourteen cases of identity theft were poorly classified as copyright infringement and two copyright infringement were improperly filed as identity theft and hacking infringement were improperly identified as theft. Number of unspecified events where test size is 0.5. It shows that the five patent attacks were poorly classified as identity theft attacks and single identity theft attacks were unfairly classified as patents.
Thanks For Reading…😊