Abstract: New and effective approaches for the analysis of seismic data make it possible to identify the distribution of earthquakes helping further to assess frequency of occurrence any associated risks. This paper proposes an effective approach for detecting areas with increased spatial density of seismic events and zoning territories on the map based on the Density-based Spatial Clustering of Applications with Noise algorithm (DBSCAN algorithm). The validity of the choice of this clustering algorithm is explained by the fact that the DBSCAN algorithm can detect clusters of complex shapes including geographical coordinates. This study uses seismic data from the seismic catalog of the Republic of Kazakhstan from 2011 to 2021 inclusive. Finally, the clusters detected over a certain period of time allowed for the presentation of a spatial model of the distribution of earthquakes and the detection of areas with increased spatial density on the map. In general, the results of the study were also compared and well associated with the general map of the seismic zoning of the Republic of Kazakhstan showing reliable results of clustering based on density. In addition, the architecture of intelligent information and the analytical system for analyzing seismic data is based on the proposed approach.

Keywords: data mining; machine learning; data analysis; clustering; DBSCAN algorithm; DBSCAN algorithm’s parameters determination; intelligent information and analytical system

Abstract: Intelligent Transportation Systems and particularly vehicular adhoc networks (VANETs) play a key role in enabling Smart Cities as well as improving and maintaining road safety. VANETs are distributed networks built from moving vehicles on the road. Each vehicle of the network has an embedded IEEE 802.11p interface to support the interaction between the vehicles and their environment (V2X) and enable Inter Vehicular Communication (IVC). However, due to the instable nature of these networks caused by the high-speed mobility of the vehicles as well as frequent fragmentation and disconnection of the network, it is necessary to design and implement robust and fault tolerant communication protocols especially in the case of emergency situations on the road to rapidly alert the environment and the competent authorities. Moreover, the communication in these networks suffers from limited bandwidth spectrum making information dissemination time critical to achieve fairness toward all the nodes of the network. This paper proposes an optimization of the Ad-hoc Multi-hop Broadcast (AMB) protocol for the dissemination of information and particularly Emergency messages in vehicular ad-hoc networks (VANETs). The proposed solution aims to reduce the network traffic while optimizing the communication time and achieving high reliability for the emergency messages. The performance of the proposed protocol is evaluated on theoretical considerations and numerical calculations.

Keywords: Vehicular ad hoc networks; 802.11p; V2X; Data dissemination; Road safety

Abstract: The paper is focused on examination of properties of transformations of Petri nets and process algebra specifications. After a brief introduction to formal methods and the transformations used, we provide descriptions of several experiments regarding Petri nets and process algebra transformations we accomplished using our transformation tools. The motivations behind this research are to practically evaluate the benefits we would gain by transformation in the field of analysis of resulting specification as well as to verify the accuracy and correctness of the tools by performing the transformations in both directions. By evaluating the experiments we were able to better perceive the actual state of the tools in practical level, the role of transformations in the field of formal methods integration and to collect some suggestions which may stimulate our further research in the given field.

Keywords: formal methods; Petri nets; process algebra; transformations; ACP; analysis

Abstract: This paper presents a grid-based algorithm using Gaussian Processes to predict outputs using spatially and temporally dependent data. First, independent Gaussian Processes are formulated along space and time axes. Then, these processes are coupled with a common noise in the covariance kernel. This common noise acts as a smoothing parameter, trading off accuracy at knots for extrapolation capabilities. The algorithm can predict time-series at unmeasured locations. The efficiency of the algorithm is demonstrated in a traffic flow prediction problem. Results suggest that applying a common additive noise term capturing cross covariances improves prediction accuracy when extrapolating outside the dataset.

Keywords: Gaussian Process; Spatio-temporal prediction; Traffic flow prediction

Abstract: This paper applies a Fractional Order Sliding Mode Control (FOSMC) to the two loops speed (the outer loop) and current (the inner loop), for Switched Reluctance Motor (SRM). This approach proposes a new, simple and fast switching control law for the Fractional Order Sliding Mode Control (FOSMC), characterized by its simplicity of design, flexibility of control and adaptive capability. The proposed controller is based on nonsingular terminal SM surface. The stability of the proposed approach was analyzed and guaranteed, using the Lypunov stability theory. This new scheme achieved minimum torque and speed ripples. Simulation results using MATLAB/SIMULINK validated the improved performance of the proposed approach against parameters variations, external disturbances and measurement noise, by comparing it with PI, Neural Network Controller (NNC), Hysteresis Controller (HC) and conventional Sliding Mode Controllers (SMC).

Keywords: Fractional order SMC; Speed Control; Current Control; Switched Reluctance Motor (SRM); Neural Network Controller (NNC)

Abstract: Cryptocurrency is the blockchain financial technology used for transactions in financial institutions and exchanges. Bitcoin has attracted much coverage from investors and commentators as it represents the maximum market capitalization on a crypto-currency exchange. The study aims to determine the correlation between the daily log-returns and to understand the tendencies in the cryptocurrency market instability of Bitcoin, Litecoin, XRP, Nxt, Dogecoin, Vertcoin, DigiByte, DASH, Counterparty, and MonaCoin. The correlation among the selected cryptocurrencies exists in the study. The analysis is focused primarily upon reference information from the preserved servers of cryptocurrency websites and finance.yahoo.com. This research assesses regular details on the Logarithmic return of Bitcoin, Litecoin, XRP, Nxt, Dogecoin, Vertcoin, DigiByte, DASH, Counterparty, and MonaCoin for a timeframe spanning from October 01st, 2014, to April 30th, 2020. From 131 cryptocurrencies, we considered only 10 Cryptocurrencies due to the availability of data after October 2014. Where there was insufficient information, there were average results determined from preceding and succeeding data. Findings demonstrate that there is GARCH modelling of cryptocurrencies against Bitcoin. Litecoin, XRP, Nxt, Dogecoin, Vertcoin, DigiByte, DASH, Counterparty, and MonaCoin; variability values throughout the duration had a significant effect on the updates from Bitcoin returns. We believe that it helps create information and resources that are valuable to practitioners and scholars who research and form cryptocurrency markets in the future.

Keywords: Blockchain; Bitcoin; Counterparty; Cryptocurrency; Dash; DigiByte; Dogecoin; Litecoin; MonaCoin; Vertcoin; XRP

Abstract: Computer-aided diagnosis (CAD) is an effective resource for diagnosing brain disorders rapidly and is also used for reducing human diagnostic errors to enhance and extend the quality of patient life. The deep learning model can self learn and generalize over a huge volume of data, it has recently gained a lot of interest over the research community in classifying medical images. But deep learning model created from the scratch takes more training time as well as a huge amount of data. Using pre-trained networks for a new, similar problem is the fundamental idea of transfer learning. In this work, the survey on disease diagnosis using deep learning-based transfer learning with Brain MRI images alone is carried out over the last 5 years. The inference drawn from this work is that a hybrid architecture based on transfer learning produced more than 90% accuracy in most of the cases with minimal training time. In hybrid architecture, more than one pre-trained models are integrated to extract high-level features. Pre-trained models are good at recognising high-level features like edges, patterns, and so on. The model designed with pre-trained model starts with learned weights rather than assigning a random value. This promotes faster convergence and, as a result, reduces the amount of time required to train the model.

Keywords: Deep learning; Brain MRI; Convolution Neural Network; Deep neural network; Transfer Learning

Abstract: The goal of this work is to automatically analyze the emotional status of speakers, in human-human interactions, carried out in TV debates, where controversial topics are often presented. Human observers provide their perception about the emotional status associated to the interventions of the participants. An analysis of the resulting annotation was carried out by using different models for representing the emotions. The obtained labeled corpus was used to build an automatic system capable of detecting the emotional status associated to each acoustic signal, making use of the deep learning paradigm. The use of a corpus, where the real emotions that appear in a Spanish TV debate (with subtleties and often closer to neutrality than acted ones), are represented is crucial for learning models properly. In fact, although the level of accuracy depends on the problem complexity and the model employed for representing the emotional status, F1 scores of 0.7 were attained.

Keywords: emotion detection; human-human interaction; speech; behavioral analysis

Abstract: In studying the processes of social influence, charisma is a matter of great importance for establishing a persuasive dialogue. Psycho-social research focused primarily on classical political media while neglecting the charisma conveyed by leaders through processes of self-presentation within computer-mediated interactions (i.e. social media). The first study analyses the charisma conveyed and the emotions expressed by Italian political leaders of three different orientations (center-left, center-right and no ideological orientation) from a multimodal perspective, considering both verbal and bodily aspects, pointing out three different "charismatic", self-presentation strategies in their Facebook videos and pictures. Results highlight how political and power positions can differentiate the type of charisma displayed and the emotions expressed in online settings. Furthermore, a second perceptual study tested the potential followers’ emotional and evaluative processes, showing how expressing negative emotions causes more negative emotions (sadness and bitterness) and evaluations (false, astute and dangerous) within the social media users perception, whereas an emotionally regulated charismatic politician can increase the users’ perception of their competence and thereby, raise their voting intentions.

Keywords: charisma; multimodal charisma; social media communication; political emotions; self-presentation strategies

Abstract: The detection of cognitive problems, especially in the early stages, is critical and the method by which it is diagnosed is manual and depends on one or more specialist doctors, to diagnose it as the cognitive decline escalates into the early stage of dementia, e.g., Alzheimer's disease (AD). The early stages of AD are very similar to Mild Cognitive Impairment (MCI); it is essential to identify the possible factors associated with the disease. This research aims to demonstrate that automated models can differentiate and classify MCI and AD in the early stages. The present research used a combination of Machine Learning (ML) algorithms to identify AD, using gene expressions. The algorithms used for the classification of cognitive problems and healthy people (control) were: Linear Regression, Decision Trees (DT), Naîve Bayes (NB) and Deep Learning (DP). The result of this research shows ML algorithms can identify AD, in early stages, with an 80% accuracy, using a Deep Learning (DL) algorithm.

Keywords: Machine Learning; Alzheimer disease; Mild Cognitive Impairmen; Deep Learning

Abstract: This paper examines recent studies within a general project, aimed at writing down the lexicons of a Conductor’s body signals. Three observational studies focusing, respectively, on the semantic area of intensity and on the modality of gaze, through annotation of several fragments of orchestra and choir conduction in concert and rehearsal, discovered a total of 100 signals of intensity in various modalities, out of which, 21 intensity gestures and 20 conducting items of gaze. Subsequent perception studies on a subset of these signals reveal slight differences in interpretation by music Experts, Non-Experts and Amateurs, but confirm their comprehensibility, probably due to their underlying semiotic devices, the same holding also in everyday life gestures and gaze.

Keywords: Multimodality; lexicons; gesture; gaze; music; Conductor

Abstract: International cross-border co-operations can highly contribute to individual and even regional value-added generation. In the present work 51 Central European stakeholders with different organizational types were selected for an online survey where preference options in a form of a multiple choice questionnaire were collected on their interests for a possible future co-operation and additional textual data on their general descriptions, strengths, focuses, goals etc. Based on the gathered data clustering of the partakers were performed with the K-modes algorithm for the categorical variables resulted of the questionnaire. Additionally Latent Dirichlet Allocation was used for the textual information in order to present an alternative technique for decision makers to the grouping of business organisations. Such tools can aid more effective and long-term business network formations and contribute to business sustainability and resilience that is of paramount importance in a globalized market framework. This globalized, turbulent environment poses risks that are hardly manageable on an individual level and therefore collaborations and augmentation of business activities gain more and more importance and focus from the side of policy makers and scientific community as well.

Keywords: cross-border co-operation; K-modes; resilience Latent Dirichlet Allocation; topic modeling;