previous day
all days

View: session overviewtalk overview

09:00-10:30 Session 18: Plenary Session: Tutorial

We regret to inform that this tutorial is cancelled as the tutor can not attend ICTERI 2018 due to personal reasons.

Location: Assembly Hall
Blockchain technology and applications
SPEAKER: unknown

ABSTRACT. A blockchain is a public ledger to which everyone has access but without a central authority having control. It is an enabling technology for individuals and companies to collaborate with trust and transparency. One of the best know applications of blockchains are the cryptographic currencies such as Bitcoin and others, but many other applications are possible. Blockchain technology is considered to be the driving force of the next fundamental revolution in information technology. Many implementations of blockchain technology are widely available today, each having its particular strength for a specific application domain. The workshop provides the participants with insights and practical experience on Blockchain technology and applications in practice, as well as theory based exploration of possible business cases.

09:00-10:30 Session 20A: Information Systems: Technology and Applications
Location: 18
Conflict control of spreading processes on networks

ABSTRACT. The focus of this work is to provide introduction to the current state of art of the field of spreading processes on networks in connection with optimal control theory and game theory. This is challenging problem which remains open, so we present problem formulation, make suggestion of possible ideas of solution and show simulations to substantiate these ideas.

Collecting the Seminal Scientific Abstracts with Topic Modelling, Snowball Sampling and Citation Analysis

ABSTRACT. This paper presents a complete information technology for collecting and analysis of a citation network of scientific publications aimed at detecting of seminal papers in a selected domain of research. The technology consists of the seed paper selection, plain snowball sampling, probabilistic topic modeling, greedy restricted snowball sampling, and analysis of the collected citation network. The topic model is built on the base of word-word co-occurrence probability with combination of sparse symmetric nonnegative matrix factorization and principal component approximation. Experiments with the collection of High Energy Physics abstracts show that the number of topics in the model is determined in natural way and the Kullback-Leibler divergence correlates with cosine similarity calculated from keywords provided by publication authors. The citation networks on "critical thinking" and "automatic pronunciation assessment" domains are collected and analyzed. The analysis shows that both the citation networks are "small worlds" and therefore the observed saturation of the restricted snowball sampling can provide the complete set of publications in domains of interest. Multiple runs of the sampling confirm the hypothesis that the set of seminal publications is stable with respect to variations of the seed papers. The modified main path analysis allows to distinguish the seminal papers including new publications following main stream of research.

Optimization of Parallel Software Tuning with Statistical Modeling and Machine Learning

ABSTRACT. High-performance computation is the main goal of parallel computers, but the performance of compiled code is often far from the best. Parallel program auto-tuning is the method adjusting some structural parameters (mainly, data structures) of an application program for a target hardware platform to speed-up computation as much as possible. In previous work, the authors have developed a framework intended to automate generation of an auto-tuner from an application source code. However, auto-tuning for complex and nontrivial parallel systems is usually time-consuming due to empirical evaluation of huge amount of parameter values combinations of an initial parallel program in a target environment. In this paper, we propose to improve the auto-tuning method using statistical modeling and neural network algorithms that allow to reduce significantly the space of possible parameter combinations. The resulting optimization is illustrated by an example of tuning a parallel sorting program, that combines several sorting methods. The optimization is done by means of the automatic training of a neural network model on results of “traditional” tuning cycles with subsequent replacement of some auto-tuner calls with an evaluation from the statistical model.

10:30-11:00Coffee Break
11:00-12:30 Session 19A: Advances in ICT Research
Location: 112
Ukrainian Banks' Business Models Under Systemic Risk

ABSTRACT. In this article we analyze specific origin of business models in Ukrainian banking system over the period of 2014-2017. Using K-mean clustering techniques five basic business models were identified due to the combination of bank asset items and liabilities sources, retail and corporate focus, equity to assets ratio which appears to be abnormally high for frozen banks. We produced migration matrix for business models from the start of systemic crisis in 2014 till recovery in 2017. We analyzed how defined strategies have affected risk and efficiency of Ukrainian banking system under systemic instability. The results of the study contribute to a deeper understanding of riskiness of business models through different periods of financial cycle. Retail and particularly "non-scheme" corporate bank business models were the most sustainable compared with "retail funding to corporate lending" type of banks. Our results enable to develop more efficient macroprudential tools grounded on heterogeneity of bank business strategies.

Plausible Event-Tree Networks for Knowledge Representation in Real-Time GIS-Based Decision Support Systems

ABSTRACT. Event-based knowledge representation models providing sufficient detalization in space and time are often necessary for real-time GIS-based decision support sys-tems (DSS). The paper is devoted to developing such model based on a plausible event tree network (PETN). PETN is built over a spatial model of a terrain discre-tized with a grid of uniform-sized cells. Each event has not only time reference, but also spatial reference, and describes a transition of the cell from one state to another. It combines different kinds of likelihood assessments (probabilistic, fuzzy, or rough) using various plausibility models. The paper describes the PETN-based knowledge representation, which can be used to describe a multitude of interacting processes on the terrain. An experiment based on real data describing a forest fire cascade was conducted and confirmed the validity and usability of proposed PETN model for the considered class of GIS-based DSSs.

Computerized intelligent system for remote diagnostics of level sensors in the floating dock ballast complexes

ABSTRACT. In this work the development of a specialized computerized system for remote diagnostics of level sensors of floating dock ballast system is presented. Ballast system of floating dock and requirements for reliable measurement of fluid levels in ballast tanks are descripted in detail. The hierarchical functional structure of the proposed remote diagnostic system consists of a multiprocessor computing complex with the corresponding software and the branched structure of digital devices. The authors propose a method of checking the correctness of level sensors that generally increases system reliability. The diagnostic calculations of the measurements correctness of the level sensors are performed on the basis of programmable logic device (PLD) with the Field-Programmable Gate Array (FPGA) architecture. The collection of diagnostic information from PLD is processed by a single-board computer that transmits data via the Internet to the cloud service “ThingSpeak”. The overall results of work of the remote diagnostics system for level sensors are displayed graphically in real time on any, specialized for these tasks, computer or mobile device that has Internet access.

11:00-12:30 Session 19B: ICT in Education
Location: Assembly Hall
ICT Using in Integrated Teaching Management Core Courses in a Foreign Language

ABSTRACT. The paper deals with the practical importance of using ICT in education which is difficult to overestimate. The research studies the process of integrated teaching management core courses in English using ICT with the purpose of developing the technology of integrated building of professional and foreign language competences using ICT. The authors analyze the ways of the effective ICT use in the integrated teaching the courses “Corporate Social Responsibility” (in English) and “English for professional purposes” and statistically prove their effectiveness. The developed integrated model using ICT has proved the synergy effect al-lowing the effective acquisition of professional information on the one hand and the development of both professional and foreign language skills on the other hand, what ensures building the professional competence as a whole. The results of the experiment can be used in developing of the recommendations as to using ICT in the integrated teaching of core courses and foreign language to future managers.

Identification of IT Sector Stakeholder’s Requirements to Masters Program in Information System in Lviv Region
SPEAKER: Ivan Izonin

ABSTRACT. The overall objectives of the study were to research the competences required by the graduates of master program in Information System (IS) to be effectively employed by the IT companies and to develop master program in Information System (IS) reflecting those requirements. Therefore, we conducted the survey among the main stakeholders of IT sector (employers and graduates of IT specialists) to identify their requirements to the IS program. The results of the survey appeared to be different for various Ukrainian regions. To be more specific, the requirements to IS program for Kherson region (mostly managerial skills) are different from the requirements to IS program in Lviv region (both technical and managerial skills). According to Tuning approach, the results of the survey were used as the ground for the development of master’s program in Information System. As the outcome of the survey the core courses, which should be the part of IS program in every University participating in the project, were determined and elective courses which cover competences required in the specific region were selected. The findings of the study would be useful both to the Universities delivering master’s in IS and to the practitioners in order to respond to the labor markets needs.