Using machine learning algorithms to detect frauds in telephone networks

This paper presents an analysis of voice traffic in telephone networks, based on machine learning algorithms to detect frauds made by callers. Starting from the raw data set that includes information about the call date, destination number, duration and caller's number, in our approach we were able to identify fraudulent calls in early stages. For balance, the data set was split in 2 parts: one for training and one for testing. To obtain mean’s values from dataset, a standardization technique was applied in order to scale the data before the dimensionality reduction using Principal Component Analysis. Then, the first two components were used as inputs for Logistic Regression and Random Forest models, having the caller as target. Finally, the target was moved on the destination file so as to identify the caller and the moment when the call has started based on a vector representation of words.

Classification overview of the Hardware Trojans in digital circuits

This paper presents an overview of the Hardware Trojans classification methods. A malicious entity can introduce a Hardware Trojan (HT) into a design in order to denial of service, destroy or disable the system. Moreover, it could leak the confidential information and the secret keys before altered them. The Hardware Trojan (HT) threats should be analyzed with maximum importance through the entire lifecycle of the integrated circuit (ICs). A hardware protection against the detected harmful logic should also be implemented.

The Robots with soul and the use of the ontologies in the human-robot communication

Robots with soul become an important field of robotics and have several issues. An interesting issue is their ability to perceive human emotions differently as the human brain. In order to recognize the human emotions a robot with soul needs to focus on Artificial Intelligence and an Artificial Neural Network, recognizing the words, the languages and the specific patterns and identifying the context and the user’s intention. In this work we propose a robot with soul, which uses an emotional ontology – based Artificial Neural Network, created in Matlab 2019, offering to the humans more privacy, a personalized conversation and a new emotion’s achievement.

Prediction error's minimization through a controller based on a Metaheuristic Algorithm

A Metaheuristic Algorithm (MA) can be a realistic method to solve a given Optimal Control Problem (OCP), but the result is an open-loop solution. If the Metaheuristic Algorithm is integrated within the Model Predictive Control (MPC) structure, a closed-loop solution can be achieved. The controller works using a prediction technique and prediction error's minimization. On the other side, the MA optimizes (minimizes or maximizes) the OCP's objective function. The controller is faced with two optimization tasks. This paper proves through theoretical analysis and simulations that the prediction error's minimization is implicitly accomplished.