Time insertion functions as a way how to guarantee state-based security with respect to timing attacks are proposed and studied. As regards the security property, we work with the property called process opacity. First, we define timing attacks and later we show how security with respect to them can be enforced by such functions. The time insertion function can alter the time behaviour of the original system by inserting some time delays to guarantee its security. We investigate conditions under which such functions do exist and also some of their properties.
A dataset containing over 13k samples of dry beans geometric features is being analysed using machine learning (ML) and deep learning (DL) techniques with the goal to automatically classify the bean species. First the original dataset was reduced to eliminate redundant features (too strongly correlated and echoing others). Then the dataset was visualised and analysed with machine learning techniques: Multinomial Bayes, Support Vector Machines, Decision Tree, Random Forest, Voting Classifier and Artificial Neural Network. The overall accuracies obtained were in range: 88.35 – 93.61%.
To a certain extent, humans and many other biological agents are able to anticipate the consequences of their actions and adapt their decisions based on available information on current and future states of their environment. The same principle can be applied to enable decision-making in artificial agents. In order to decide on an action, an agent could envision the consequences for each of the actions and then choose the one promising the best outcome. This anticipatory scheme can enable fast decisions in highly dynamic and complex situations, which has been demonstrated in humanoid robots playing soccer. We extend this principle to the scenario of bio-hybrid beehives augmented with robotic actuators, which allow to influence the foraging locations of the bees. We investigate how a bio-hybrid beehive can make decisions and direct the bees in a way which would benefit the the whole ecosystem enabling sustainable beekeeping. We explore the general principles of anticipation and discuss connections to cognitive science and developmental robotics. We present an implementation of a simulator for the behavior of the augmented beehive and present preliminary results demonstrating the feasibility of the anticipatory approach.
Formal definitions of state-based and language-based security with respect to timing attacks are proposed and studied. Then various ways how to secure systems with respect to such attacks are discussed. First, we investigate time insertion functions. Conditions, when such functions exist and could protect systems, are investigated. Then we discuss the concept of supervisor control which can be used if there is no appropriate time insertion function to protect the systems.
Recommender systems play a key role in many branches of the digital economy. Their primary function is to select the most relevant services or products to users’ preferences. The article presents selected recommender algorithms and their most popular taxonomy. We review the evaluation techniques and the most important challenges and limitations of the discussed methods. We also introduce Factorization Machines and Association Rules-based recommender system (FMAR) that addresses the problem of efficiency in generating recommendations while maintaining quality.
