# Processing with AI
## Exploration : IA & Ethics

Name:
>Studetector
>
Subject:
>Detect student presence in class using Face Recognition

>[TOC]

## Design brief

### Biais

If we don't source our dataset with enough rigor, the following biais might appear:

>1. Make mistakes between students by confusing students who look alike
>2. Create a kind of "fear" if the youngsters learn about the process of face recongnition at their workplace

We will ensure that our model is not biaised by:

>1. Sourcing our data from the finest machine learning code existing in order for it to best recognize which student is present or not with the less error possible
>2. Making sure our data take into account facial recognition whith the most precise face point detection to avoid confusing twins for example
>3. The infrared dot projection as used on the iPhone X is a good way to get the student's face mapped as well as possible

### Overfitting

We will make sure our model does not overfit by
> Checking the accuracy of our model on several models and training it until the detection accuracy exeeds 99%

### Misuse

>We have to remind ourselves that our application could be misused by cyber pirates if the software gers hacked. So we must build a firewall that takes into account the possibility of a cyber attack

### Data leakage

>In a catastrophic scenario, where all of our training dataset were stolen or recovered from our model, the risk would be the publishing of the student list and personal data or the selling of the latter to a bad intended organization.

### Hacking

> If someone found a way to "cheat" our model and make it make any prediction that it want instead of the real one, the risk would be that students would never come back to class and it would trick the student participation/presence grades