OVERVIEW OF NETWORK-BASED METHODS FOR ANALYZING FINANCIAL MARKETS
Pavel Tsankov

Abstract
Network based methods are suitable for the analysis of a large number of financial time series and a better understanding of their interdependencies. Known approaches to reveal the underlying information about the complex structure of these interdependencies include network-wise and vertex-wise measures of the topology, as well as filtering techniques relying on minimum spanning trees, planar graphs, or spectral analysis. The aim of this study is to review relevant graph theoretical and statistical models and techniques for generating and examining the properties of financial networks, obtained by computing time series correlations or causality relationships. In particular, this study reviews literature discussing the time evolution of the observed phenomena from a network perspective, as well as applications in economy and finance, ranging from risk and diversification, through policy making and better understanding crisis impact, to forecasting. The information synthesized in this paper can be useful to gain further insights into this relatively new research area.

Keywords
causality, degree stability, financial networks, time-varying graphs, topology.

Full Text:
PDF
                                                                                                                                                                                                                                                             Top
DOI: 10.47978/TUS.2021.71.01.001

WIDGET TRANSACTIONS DYNAMIC PROCESSING ARCHITECTURE DEVELOPMENT
Viktoriia Merlak, Daryna Hrebeniuk, Galina Cherneva

Abstract
The article is devoted to the development of a widget's dynamic transaction processing architecture. An object is a hierarchical widget in a user interface organization. The subject is the process of the dynamic processing of widget transactions. The purpose of the article is to develop a dynamic transaction processing architecture) for a hierarchical widget. The article discusses the construction of a complex structured user interface in applications that are built on the basis of event-oriented programming. The article discusses the use of hierarchical widgets in companies such as Microsoft and Google, as well as existing developments related to hierarchical widgets. The main components of the dynamic processing of widget transactions, the main objects of the dynamic model are determined. A hierarchy of objects of the dynamic model of transaction processing of the widget is proposed. To interact with the dynamic model and its graphical representation, a special graphical notation is proposed.
Dynamic processing of widget transactions is a new approach that has its advantages and disadvantages, the class of tasks being solved. The proposed architecture combines both work with unstructured data and the use of a hierarchical data model.

Keywords
dynamic processing, hierarchical widget, transaction.

Full Text:
PDF
                                                                                                                                                                                                                                                             Top
DOI: 10.47978/TUS.2021.71.01.002
 

DEVELOPMENT OF AN INFORMATION MODEL FOR THE PERSONALITY’S SOCIAL PORTRAIT FORMATION USING OSINT TECHNOLOGY
Mykhailo Mozhaiev, Pavlo Buslov

Abstract
The results of the development of an information model for the personality’s social portrait formation are presented. The modelling has been carried out using OSINT technology that is the technology of legal obtaining and using open source information.
In the result of the analysis, it has been found out that the social portrait is a heterogeneous semantic network consisting of personalized data. It has been defined that people organize formal and official communities of various orientations and the number of such communities associated with a particular person is practically unlimited.
When formalizing the decision-making process, the concept of a group social portrait (GSP) has been introduced, which takes into account the community’s social tendencies united by certain common properties, group members' interpersonal interactions and their behavioural patterns.
The obtained information models of personal and group social portraits let to take into account all the main properties of the objects under study, their tonality, and significance, as well as to conduct an analysis of the implicit dependencies determination. The next step is to move on to considering the diversity of the digital social environment elements.

Keywords
decision-making support systems, information model, information technology, open-source intelligence, social portrait.

Full Text:
PDF
                                                                                                                                                                                                                                                             Top
DOI: 10.47978/TUS.2021.71.01.003

 

SPECIFICS AND VULNERABILITIES OF THE TIMING CONTROL IN CYBER-PHYSICAL SYSTEMS
Iliya Georgiev, Ivo Georgiev

Abstract
Cyber-physical systems integrate powerful computing (real-time embedded system, operating system, applications, and Internet networking) and physical environment (advanced manufacturing cells, medical platforms, energetics aggregates, social and educational control). The reliable functionality depends extremely on the correct timing. Wrong timing because of buried malfunction or external tampering could be critical. The paper is some analysis of the vulnerable timing parameters that influence the precise processing. Expert estimation of the criticality of different timing parameters is given to support fault-tolerant design considering possible failures.

Keywords
cyber-physical systems, real-time, timing control, vulnerabilities, Internet of Things
Full Text:
PDF

DOI: 10.47978/TUS.2021.71.01.004

 

A NOVEL BIOLOGICALLY INSPIRED DEVELOPMENTAL INDIRECT ENCODING FOR THE EVOLUTION OF NEURAL NETWORK CONTROLLERS FOR AUTONOMOUS AGENTS
Stefan Tsokov, Milena Lazarova, Adelina Aleksieva-Petrova

Abstract
Evolutionary algorithms provide the ability to automatically design robot controllers, but their wider use is hampered by a number of problems, including the difficulty of obtaining complex behaviors. This paper proposes a biologically inspired indirect encoding method for developing neural networks that control autonomous agents. The model is divided into three stages, the first two stages determine the structure of the network – the positions of the neurons and the network connectivity, and the third stage, occurring during the lifetime of the agent, determines the strength of connections based on the network activity. The model was tested experimentally by simulating an agent in an artificial environment, and the results of these simulations show that the method successfully evolved agents, capable of distinguishing between several types of objects, collecting some while avoiding others, without the use of a complex fitness function.

Keywords
artificial nervous system, autonomous agent, developmental representation, indirect encoding, neuroevolution
Full Text:
PDF

DOI: 10.47978/TUS.2021.71.01.005

 

A NEURAL NETWORK WITH HFO₂ MEMRISTORS
Stoyan Kirilov, Ivan Zaykov

Abstract
In the last twenty years, the neural networks are under intensive analyses. One of the main ideas of the scientists is to partially replace some of their CMOS-based elements by memristors. Memristors are preferred for application due to their memory effect, low power consumption and nano-size dimensions. The purpose of this paper is to propose an analysis of a feed-forward neural network with HfO₂ memristor-based synapses for XOR logic function emulation. The considered network uses synaptic devices with a memristor, resistor, and a differential amplifier. The proposed synaptic scheme can ensure positive, zero and negative synaptic weights. For the neural network analysis several classical and modified HfO₂ memristor models are used. The network is successfully tested in LTSPICE. The occurrence of convergence problems is reduced by replacing the standard step function in the models by its smooth and differentiable analog. The capability of the modified models for operation in complex schemes is proven.

Keywords
neural network, memristor synapse, hafnium dioxide, memristor model, step-like logistic function
Full Text:
PDF

DOI: 10.47978/TUS.2021.71.01.006

 

APPLICATION OF NEURAL NETWORKS FOR TIME SERIES ELECTRICAL CONSUMPTION FORECAST
Verica Babamova-Tsenova, Ognyan Andreev and Georgi Tsenov

Abstract
With the introduction of IoT there is possible to use nowadays cheap measurement devices that can store information form actuators and store it in databases locally or online. With the recent opening of the electrical distribution markets, it will be possible to buy electrical energy on varying prices and from various suppliers. With big time intervals of data collection for the plant consumption and of the market price fluctuations done with IoT devices for relatively small time intervals on the recorded samples a user, electrical power profile can be created by using neural networks as time series forecasts predictor. With such a plant/user power profile, the predicted production can be shifted towards the intervals with cheaper electricity prices leading to reduced production costs. This paper presents the results for creating an energy consumption profile with electrical loads forecasts when they are presented as time series and by using MATLAB’s Neural Networks Toolbox.

Keywords
Neural Networks, Time series prediction, Electric Load Optimization
Full Text:
PDF

DOI: 10.47978/TUS.2021.71.01.007