Cyber Crime Cases by False Positive

In this article, I am going to tell about how security with ML might be dangerous in some cases.

So before going ahead what’s the case which is very dangerous let’s discuss.

We know nowadays we use AI everywhere, AI is very common, AI has lots of benefits and drawbacks.

So in this post, I am going to discuss mainly the drawbacks.

We know classification means either pass or fail, exist or not this kind of decision known as Classification, Same like in ML we have to classify the various and attack,

To classify in ML we have lots of blogs but the motto is the same after the model created when ML tries to detect various they have to decide, but the point here is that how ML classify, we classify by using a confusion matrix.

Let’s discuss the confusion matrix

Confusion Matrix -:)

A confusion matrix is a way to find the accuracy of the model and take a decision on what to do and lots more things we do in the AI world.

There are two types of error in Confusion Matrix

Type I Error -: It is very dangerous (False Positive)

Type II Error -: It is also dangerous

Let’s understand some vocab

True Positive (TP) -: ML predicted True & Actual value also Positive

True Negative(TN) -: ML predicted True & Actual value Negative

False Negative(FN) -: ML Predicted false & Actual also Negative

False Positive(FP) -: ML predicted false & its actually positive

Now Let’s discuss the cybercrime by False Positive

Today we know we want self-driven cars, In the US a self-driving car which is fully engaged with ML, a hacker attack that car and changes the traffic signal detection, the hacker changed the car signal detection, in this case, lots of people get an accident.

Keep Sharing Keep Learning


ARTH Learner | AWS | BigData | ML/AI | Ansible | Docker | CKA & CKAD | Python | Linux | RH-294