Abstract: The (benign paroxysmal positional vertigo) BPPV is the most common type of vertigo, influencing the quality of life to considerable percentage of population after the age of forty (25 out of 100 people are facing this problem after 40). The semicircular canals which are filled with fluid normally act to detect rotation via deflections of the sensory membraneous cupula. We are modeling human semicircular canals (SSC) which considers the morphology of the organs and the composition of the biological tissues and their viscoelastic and mechanical properties. For fluid-structure interaction problem we use loose coupling methodology with ALE (Arbitrary Lagrangian Eulerian) formulation. The tissue of SSC has nonlinear constitutive laws, leading to materially-nonlinear finite element formulation. Our numerical results are compared with nystagmus from real clinical patient. The initial results of 3D Tool software user interface and fluid motion simulation and measurement of the video head impulse test (vHIT) with the Oculus system are presented.

Abstract: Human hearing system is very interesting for investigation. There are several parts in hearing system, but the most important parts in sense of conversion of audio signal in electrical impulse are cochlea and organ of Corti. The reason why scientists investigate mechanical behavior of human hearing system is hearing loss – a health problem that affects a large part of the world's human population. That problem can be caused by aging, as consequence of mechanical injuries or some disease or even can be congenital. The experimental auditory measurements provide information only about the level of hearing loss, but without information what is happening in the hearing system. Therefore, it is very helpful to develop a numerical model of the parts of hearing system such as cochlea and organ of Corti to demonstrate process of conversion of acoustic signals into signals recognizable by human brain. Two numerical models are developed to investigate hearing problems: tapered three-dimensional cochlea model and twodimensional cochlea cross-section model with organ of Corti.

Abstract: In order to provide more advanced and intelligent energy management systems (EMS), generic and scalable solutions for interfacing with numerous proprietary monitoring equipment and legacy SCADA systems are needed. Such solutions should provide for an easy, plug-and-play coupling with the existing monitoring devices of target infrastructure, thus supporting the integration and interoperability of the overall system. One way of providing such holistic interface that will enable energy data provision to the EMS for further evaluation is presented in this paper. The proposed solution was based on the OpenMUC framework implementing a web-service based technology for data delivery. For testing the system prior to an actual deployment, data simulation and scheduling component was developed to provide a low-level data/signal emulation which would replace signals from the field sensors, monitoring equipment, SCADAs, etc. Furthermore, the proposed interface was implemented in a scalable and adaptable manner so it could be easily adjusted and deployed, practically, within any complex infrastructure of interest. Finally, for demonstration and validation purposes, the campus of the Institute Mihajlo Pupin in Belgrade was taken as a test-bed platform which was equipped with SCADA View6000 system serving as the main building management system.

Abstract: Fostering of ICT technologies in energy domain has garnered a significant degree of attention in the last few years, especially in the context of improving energy efficiency of large infrastructures. One such approach is presented in this paper and focuses on empowering the neighborhoods with advanced energy management functionalities through the utilization of ICT infrastructure devoted to smart energy metering, supervision, control and global management of energy systems. In particular, this paper reveals the conceptual system architecture, core energy modelling framework and optimization principles used for the development of the neighborhood energy management solution under the EPIC-Hub project.

Abstract: In this paper the correlation between different variables with the hourly consumption of electricity is analyzed. Since the correlation analyses is a basis for predicting, the results of this study may be used as an input to any model for forecasting in the field of machine learning, such as neural networks and support vector machines, as well as to the various statistical models for prediction. In order to calculate the correlation between the variables, in this paper the two main methods for correlation are used: Pearson’s correlation, which measures the linear correlation between two variables and Spearman's rank correlation, which analyzes the increasing and decreasing trends of the variables that are not necessarily linear. As a case study the electricity consumption in Macedonia is used. Actually, the hourly data for electricity consumption, as well as the hourly temperature data for the period from 2008 to 2014 are considered in this paper. The results show that the electricity consumption in the current hour is mostly correlated to the consumption in the same hour the previous day, the same hour-same day combination of the previous week and the consumption in the previous day. Additionally, the results show a great correlation between the temperature data and the electricity consumption.

Abstract: The application domain of Smart Grid is a matter of a great interest of scientific and industrial communities. A comprehensive model and systematic assessment is necessary to select the most suitable Enterprise Architecture (EA) development or integration project from many alternatives. In this paper, a Fuzzy Influence Diagrams with linguistic probabilities is proposed for the Smart Grid EA conceptual model and risk analysis. The proposal relies on a methodology devoted to decisionmaking perspectives related to EA and risk assessment on Smart Grid initiatives based on fuzzy reasoning and influence diagrams (FID).

Abstract: According to Serbian Environmental Agency (SEPA), more than 30% of citizens of Serbia were exposed to air that is considered not healthy in 2014. An important element in the work to reduce atmospheric pollution is involvement of the citizens, through awareness raising and other means such as adherence to pollution reduction measures. The aim of our work is to raise awareness of pollution gradients and levels by developing a user-friendly citizen’s observatory for air quality. This would allow the general public to generate and access information about air quality that is directly relevant to them – near their homes or work places. Citizen observatory is a quickly developing concept: it is a combination of citizen science approaches with an ICT infrastructure, to objectively monitor or subjectively assess the state of the environment (air quality), and to provide this information in a manner that can be useful to both individual citizens, their groups or the authorities. Monitoring techniques for air quality are numerous, and those that are suitable for citizen’s own measurements are just starting to be accessible. Our citizen observatory (co.citi-sense.eu) provides access to a selection of products that were developed in collaboration with users, and that allow the users to participate in air quality assessments. A practical implementation of the citizen observatory requires establishing operative connections between monitoring and data collection systems, a central information platform, and user products such as apps or web pages. Practical challenges include quality control and assurance of the input information, use of existing communication infrastructures, and not least identification of the interests and needs of the citizens: oftentimes, the awareness level of the problem addressed, and an understanding of the issues, are the limiting elements for the use of such systems. This paper shows some of the products intended for the public of Belgrade. The work is a result of the CITI-SENSE consortium efforts and can be accessed at co.citi-sense.eu*.

Abstract: The paper discusses an automation procedure for integration of Geographic Information System (GIS) and Power Network Analysis System (PNAS). The purpose of PNAS is to provide analysis and optimization of electrical power distribution system. Its functionality is based on a detailed knowledge of electrical power system model. Automation procedure problems are identified and concepts of an implemented software solution are presented. The core of the solution is the software for implementation of an internal model based on IEC 61970-301 CIM standard. Second part of the solution, which this paper is focused on, is a software layer that detects difference between two relevant GIS states just before successive synchronizations with PNAS. Using “divide and conquer” strategy, the detected difference is split into groups of connected elements. Those groups are used for update of PNAS to make it consistent with GIS. The paper also explains an algorithm for detection of the difference between two states and a procedure for formation of groups, as well as the conditions and limitations of automation procedure for incremental integration.

Abstract: The goal of this paper is, from one side, informative i.e. to introduce the existing activities in the European Commission related to metadata management in e-government systems, and from the other, to describe our experience with metadata management for processing statistical data in RDF format gained through development of Linked Data tools is recent EU projects LOD2, GeoKnow and SHARE-PSI. The statistical domain has been selected in this analysis due to its relevance according to the EC Guidelines (2014/C 240/01) for implementing the revised European Directive on the Public Sector Information (2013/37/EU). The lessons learned shortly described in this paper have been included in the SHARE-PSI collection of Best practices for sharing public sector information.

Abstract: Recent open data initiatives have contributed to opening of non-sensitive governmental data, thereby accelerating the growth of the LOD cloud and the open data space in general. A lot of this data is statistical in nature and refers to different geographical regions (or countries), as well as various performance indicators and their evolution through time. Publishing such information as Linked Data provides many opportunities in terms of data aggregation/integration and creation of information mashups. However, due to Linked Data being relatively new field, currently there is a lack of tools that enable exploration and analysis of linked geospatial statistical datasets. This paper presents ESTA-LD (Exploratory Spatio-Temporal Analysis), a tool that enables exploration and visualization of statistical data in linked data format, with an emphasis on spatial and temporal dimensions, thus allowing to visualize how different regions compare against each other, as well as how they evolved through time. Additionally, the paper discusses best practices for modeling spatial and temporal dimensions so that they conform to the established standards for representing space and time in Linked Data format.

Abstract: The paper deals with managing accreditation documents in Serbian higher education. We propose domain ontology for the semantic description of accreditation documents. The ontology has been designed as an extension of a generic document ontology which enables customization of document centric systems (DCS). The generic document ontology has been based on ISO 82045 families of standards and provides a general classification of documents according to their structure. By inheriting this high-level ontology, the ontology proposed in this paper introduces concepts for representing accreditation documents in Serbian higher education. The proposed ontology allows defining specific features and services for advanced search and machine-processing of accreditation data. An evaluation of the proposed ontology has been carried out on the case study of accreditation documents for the Software engineering and information technologies study program at the Faculty of Technical Sciences, University of Novi Sad.

Abstract: Internet social networks may be an abundant source of opportunities giving space to the “parallel world” which can and, in many ways, does surpass the realty. People share data about almost every aspect of their lives, starting with giving opinions and comments on global problems and events, friends tagging at locations up to the point of multimedia personalized content. Therefore, decentralized mini-campaigns about educational, cultural, political and sports novelties could be conducted. In this paper we have applied clustering algorithm to social network profiles with the aim of obtaining separate groups of people with different opinions about political views and parties. For network case, where some centroids are interconnected, we have implemented edge constraints into classical k-means algorithm. This approach enables fast and effective information analysis about the present state of affairs, but also discovers new tendencies in observed political sphere. All profile data, friendships, fanpage likes and statuses with interactions are collected by already developed software for neurolinguistics social network analysis - “Symbols”.

Abstract: Facebook, the popular online social network, is used on daily basis by over 1 billion people each day. Its users use it for message exchange, sharing photos, publishing statuses etc. Recent research shows that a possibility exists of determining the connection level (or strength of their friendship, tie strength) between users based on analyzing their interaction on Facebook. The aim of this paper is to explore, as a proof of concept, the possibility of using a model for calculating strength of friendship to compare and classify ego-user’s Facebook friends. A survey, which involved more than 2500 people and collected a significant amount of data, was conducted through a developed web application. Analysis of collected data revealed that the model can determine with a high level of accuracy the stronger connections of ego-user and classify ego-user’s friends into several groups according to the estimated strength of their friendship. Conducted research is the base for creating an enriched social graph – graph which shows all kinds of relations between people and their intensity. Results of this research have plenty of potential uses, one of which is specifically improvement of the education process, especially in the segment of e-learning.

Abstract: Open Data movement is gaining momentum in recent years. As open data promise free access to valuable data, as well as its usage, it has become an important concept for ensuring transparency, involvement and monitoring of certain processes. As such open data is of special importance in science and in government. At the same time yet another trend is also visible - adoption of process driven information systems. In this paper an approach to systematic enabling of open data dissemination from process driven information system is explored.

Abstract: This paper presents potentials, needs and problems of creating an adequate research area for the development of speech technologies in Serbia within the European Research Area. These technologies are a very important part of the strategy Europe 2020, because of Europe's awareness that speech technologies are one of the missing parts that will help Europe to build a unified digital market, overcome language barriers and increase the fluctuation of employees. The authors of the paper made a comparative analysis of the European actions in the European Research Area and projects in Serbia dedicated to the development of speech and language technologies. The analyses shows how European programs influence the development of speech technologies in Serbia.

Abstract: This paper addresses some issues of generative programing. The main aim of this paper is to present a DSL designed for developing web applications. It will be possible to easily generate whole web application by defining its functionalities in proposed DSL. Also, we will present possibilities to implement such DSL in Scala language. In addition, advantages of Scala language for this purpose will be discussed.

Abstract: In this paper we present a domain specific language, named MongooseDSL, used for modeling Mongoose validation schemas and documents. MongooseDSL language concepts are specified using Ecore, a widely used language for the specification of meta-models. Besides the meta-model we present the concrete syntax of the language alongside the examples of its usage. This MongooseDSL language is a part of the developed NoSQLMigrator tool. The tool can be used for migrating data from relational to NoSQL database systems.

Abstract: The paper presents a framework for developing single page Web applications based on Java technologies and AngularJS. Unlike traditional code-generation, the frontend of this solution uses a set of built-in generic user interface components capable of adapting to any metadata provided from the backend. The goal is to quickly provide a base for building a fully functional application and to switch focus from frontend development to modeling and backend development.

Abstract: The paper presents an overview of the underlying framework that provides Java code generation based on OCL constraints used in Kroki tool. Kroki is a mockup-driven and model-driven tool that enables development of enterprise applications based on participatory design. Unlike other similar tools, Kroki can generate a custom set of business rules along with standard CRUD operations. These rules are specified as OCL constraints and associated with classes and attributes of the developed application.

Abstract: The referential integrity constraint (RIC) is one of the key constraints that exists in every data model. Current XML database management systems (XML DBMSs) do not completely support this type of constraint. The structure of an XML document can contain the information about the relationship between elements, but modern XML DBMSs do not perform automatic RIC verification during database operations execution. It is therefore necessary to implement and perform additional actions to provide a fully functional validation of the referential integrity constraint. In this paper we discuss different structures of XML documents, and their impact on the referential integrity constraint. We also present two types of its implementation in XML databases: using XQuery functions and triggers. Those techniques correspond to the characteristics of a selected XML DBMS.