malware-detection

Artificial intelligence model based on a simple perceptron, with the crucial purpose of accurately predicting whether a given mobile application is malware.

Model precision : 97.98%

Authors:

• Jansel Lopez Bouza
• Elizabeth Calzadilla Fernández

Distribution of data in a three-dimensional space, malware (yellow), goodware (purple).

Confusion matrix, number of classifications (true negative - 167, false negative - 10, false positive - 9 and true positive - 707).