Comparable Evaluation of Contemporary Corpus-Based and Knowledge-Based Semantic Similarity Measures of Short Texts
Vol. 1 No. 1 (2011): JITA – APEIRON Bojan Furlan, Vladimir Sivački, Davor Jovanović, Boško Nikolić Comparable Evaluation of Contemporary Corpus-Based and Knowledge-Based Semantic Similarity Measures of Short Texts Original scientific paper DOI: https://doi.org/10.7251/JIT1101065F Abstract Download Article PDF This paper presents methods for measuring the semantic similarity of texts, where we evaluated different approaches based on existing similarity measures. On one side word similarity was calculated by processing large text corpuses and on the other, commonsense knowledgebase was used. Given that a large fraction of the information available today, on the Web and elsewhere, consists of short text snippets (e.g. abstracts of scientific documents, image captions or product descriptions), where commonsense knowledge has an important role, in this paper we focus on computing the similarity between two sentences or two short paragraphs by extending existing measures with information from the ConceptNet knowledgebase. On the other hand, an extensive research has been done in the field of corpus-based semantic similarity, so we also evaluated existing solutions by imposing some modifications. Through experiments performed on a paraphrase data set, we demonstrate that some of proposed approaches can improve the semantic similarity measurement of short text. Keywords: semantic similarity, corpus-based, knowledge-based. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators.
Measuring the Characteristics of DG CAC Algorithm
Vol. 1 No. 1 (2011): JITA – APEIRON Goran Đukanović, Milan Šunjevarić, Nataša Gospić Measuring the Characteristics of DG CAC Algorithm Original scientific paper DOI: https://doi.org/10.7251/JIT1101060DJ Abstract Download Article PDF Users today expect email and instant messaging access, surf, video games and other services through mobile broadband access networks. In order to support this increasing data traffic, advanced resource management has to be implemented. As CAC (Call Admission Control) algorithm plays an important role in this resource management, comparing of two proposed call admission control algorithms has been done in this paper. Algorithms are tested in simulation environment, for two different periods of time. They showed expected characteristics in both 1000 and 10000 seconds periods, and newly proposed DG CAC algorithm showed better results than other algorithm, in number of handover requests, and in the way of returning resources to degraded connections. Keywords: CAC, QoS, UMTS, Wireless. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators.
Extraction of Information in the Context of Business Inteligence
Vol. 1 No. 1 (2011): JITA – APEIRON Branko Latinović Extraction of Information in the Context of Business Inteligence Original scientific paper DOI: https://doi.org/10.7251/JIT1101054L Abstract Download Article PDF Business Intelligence in the developed business systems allows better reasoning and decision making. ETL processes represent the most important processes in the system of Business Intelligence. It is about extracting, transforming, and filling a Data Warehouse with data which then transforms into data that is by its nature new and presented in a way that is meaningful and useful in an actual business organization. In conjunction with the methods of Information Extraction, knowledge is significantly expanded and given a completely new image. Intention is the collection of data that is available and processing the same in one place, regardless of whether the data was in a structural form or any other. Keywords: Business Intelligence, Data Warehouse, Information Extraction. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators.
Trends in Educational Games Development
Vol. 1 No. 1 (2011): JITA – APEIRON Miroslav Minović, Dušan Starčević Trends in Educational Games Development Original scientific paper DOI: https://doi.org/10.7251/JIT1101041M Abstract Download Article PDF In this paper we will give a literature review related to game-based education, in the first place at university, as well as the analysis of existing solutions which should enable this type of eLearning. The main topic of this research will be capacity for applying modern information technologies for developing game-based learning platform. When we chose this topic, we started form the fact that there are no applied game-based eLearning systems at universities. During analysis phase, we found that more research is needed in order to improve application of games in education. In the first place, these studies should cover listed problems: how to design educative games in order to achieve better learning effects; how to develop software tools to automate educative game development process; establish methods and techniques for knowledge and skills assessment utilizing educative games. Keywords: Game-based learning, eLearning, Games, Motivation for learning. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators.
Model for Managing Software Development Projects by Fixing Some of the Six Project Management Constraints
Vol. 1 No. 1 (2011): JITA – APEIRON Boris Todorović, Miroslav Matić Model for Managing Software Development Projects by Fixing Some of the Six Project Management Constraints Original scientific paper DOI: https://doi.org/10.7251/JIT1101033T Abstract Download Article PDF This study is focused on the software development process, viewed from perspective of information technology project manager. Main goal of this research is to identify challenges in managing such projects and provide a model for delivering software solutions that satisfies client’s expectations. Project management theory describes six constraints or variables in every project, which project managers can use to better control the project and its outputs. Fixing some of the six project management constraints (scope, cost, time, risks, resources or quality) will allow project manager to focus on most important project aspects, rather than being drawn between all of the variables.This paper is aimed at information technology project managers and portfolio managers, as it describes the practical application of this model on a software development project. Findings of this research support the theory that, by applying good project management practice and focusing on project/business-critical requirements, will enable project managers to complete projects successfully and within tolerance limits. Results show that by identifying key business constraints, project managers can create good balance of six constraints and focus on the most important ones, while allowing other constraints to move between limits imposed by clients and stakeholders. Keywords: software development, project management, PMBOK, six project constraints, fi xed project constraints, risk management, quality management, project scope management. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators.
The Usage of Information Technology in the Implementation of the Bologna Principle of the Student Mobility
Vol. 1 No. 1 (2011): JITA – APEIRON Gordana Radić The Usage of Information Technology in the Implementation of the Bologna Principle of the Student Mobility Original scientific paper DOI: https://doi.org/10.7251/JIT1101024R Abstract Download Article PDF In this paper, student mobility is observed as one of the steps in realization of the Digital Agenda of the European Union. Student mobility, as one of the main principles of the Bologna process, is the means of effectiveness increase and quality of the educational system in European Higher Education Area, EHEA, because it enables better exchange and flow of knowledge and ideas, as well as the adoption of good practices. Management Identity (IdM) system of the Higher Educational institution is a system which supports student mobility by using personal information when accessing data. The basic identity document in this system is a student smart card with the owner’s fingerprint. This biometrical data insures high level of data and identity protection. This paper proposes informational system which, in itself, contains standards for student mobility support as one of the modules of the IdM system of the Higher Educational institution. Keywords: Identity management, student mobility, smart card, biometrics. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators.
A Multimodal Approach to Design of Aircraft Cockpit Displays
Vol. 1 No. 1 (2011): JITA – APEIRON Mlađan Jovanović, Dušan Starčević, Mirko Petrović A Multimodal Approach to Design of Aircraft Cockpit Displays Original scientific paper DOI: https://doi.org/10.7251/JIT1101016J Abstract Download Article PDF In this paper, we present an approach to design of command tables in aircraft cockpits. To date, there is no common standard for designing this kind of command tables. Command tables impose high load on human visual senses for displaying flight information such as altitude, attitude, vertical speed, airspeed, heading and engine power. Heavy visual workload and physical conditions significantly influence cognitive processes of an operator in an aircraft cockpit. Proposed solution formalizes the design process describing instruments in terms of estimated effects they produce on flight operators. In this way, we can predict effects and constraints of particular type of flight instrument and avoid unexpected effects early in the design process. Keywords: Aircraft cockpit, multimodal user interfaces, aircraft instrument, formal description of cockpit display. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators.
Applications of Smartphones for Ubiquitous Health Monitoring and Wellbeing Management
Vol. 1 No. 1 (2011): JITA – APEIRON Mladen Milošević, Michael T. Shrove, Emil Jovanov Applications of Smartphones for Ubiquitous Health Monitoring and Wellbeing Management Original scientific paper DOI: https://doi.org/10.7251/JIT1101007M Abstract Download Article PDF Advances in smartphone technology and data communications facilitate the use of ubiquitous health monitoring and mobile health application as a solution of choice for the overwhelming problems of the healthcare system. In addition to easier management and seamless access to historical records, ubiquitous technology has the potential to motivate users to take an active role and manage their own conditions.In this paper we present capabilities of the current generation of smartphones and possible applications for ubiquitous health monitoring and wellness management. We describe the architecture and organization of ubiquitous health monitoring systems, Body Sensor Networks, and integration of wearable and environmental sensors. We also describe mainstream mobile health related applications in today’s mobile marketplaces such as Apple App Store and Google Android Marketplace. Finally, we present the development of UAHealth – our integrated mobile health monitoring system for wellness management, designed to monitor physical activity, weight, and heart activity. Keywords: Smartphone, Body Sensor Networks, Health, Wellbeing. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators.
Monitoring of Jee Applications and Performance Prediction
Vol. 1 No. 2 (2011): JITA – APEIRON Dušan Okanović, Milan Vidaković, Zora Konjović Monitoring of Jee Applications and Performance Prediction Original scientific paper DOI: https://doi.org/10.7251/JIT1102136O Abstract Download Article PDF This paper presents one solution for continuous monitoring of JEE application. In order to reduce overhead, Kieker monitoring framework was used. This paper presents the architecture and basic functionality of the Kieker framework and how it can be extended for adaptive monitoring of JEE applications. Collected data was used for analysis of application performance. In order to predict application performance, regression analysis was employed. Keywords: continuous monitoring, Java, JMX, regression analysis. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators.
The Way of Students’ Efficiency Improvement in Knowledge Acquisition and Transfer Knowledge Model in Clarolina CMS
Vol. 1 No. 2 (2011): JITA – APEIRON Nevzudin Buzađija The Way of Students’ Efficiency Improvement in Knowledge Acquisition and Transfer Knowledge Model in Clarolina CMS Original scientific paper DOI: https://doi.org/10.7251/JIT1102127B Abstract Download Article PDF In this work, throughout the research which was organized in one high school in Bosnia and Herzegovina, it will be shown the influence of exercises on the final result in the e-learning environment at the final test done by students. The research was conducted from the subject informatics in the I, II and III grade. The type of the questions were of multiple choices, addition and accession. The aim was to see how much influence these online exercises have on the final outcome which is demonstrated through the final informatics test done by students and which is done in a classical way in classroom after the finished teaching materials that were planned according to high school rules. In the research, it was taken account of making all preconditions available for easy experiment conducting with regard to technical securing preconditions for students access to blended system of teaching. Concerning the recent experience, it is noticeable that youth like the use of IT and communication devices. In order to secure all necessary conditions, it was conducted the survey among students about having technical preconditions of online access to testing and about students knowledge of work principle in the Claroline LMS platform. The aim was to increase motivation of high school students with regard to the use of online materials, because in high schools of Bosnia and Herzegovina almost nothing is undertaken when it comes to the implementation of new IKT possibilities. Keywords: knowledge transfer, blended learning, Claroline, e-learning, exercises, motivation and web technology. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators. Vol. 26 No. 2 (2023): JITA – APEIRON Igor Shubinsky, Alexey Ozerov Application of Artificial Intelligence Methods for the Prediction of Hazardous Failures Original scientific paper DOI: https://doi.org/10.7251/JIT2302061S Download Article PDF Abstract The availability of real-time data on the state of railway facilities and the state-of-the art technologies for data collection and analysis allow transition to the fourth generation maintenance. It is based on the prediction of the facility functional safety and dependability and the risk-oriented facility management. The article describes an approach to assessing the risks of hazardous facility failures using the latest digital data processing methods. The implementation of this approach will help set maintenance objectives and contribute to the efficient use of resources and the reduction of railway facility managers’ expenditures. Keywords: predictive analysis, maintenance, functional safety, Big Data, Data Science, risk indicators.