Papers by Vladimir Radevski
2023 IEEE International Mediterranean Conference on Communications and Networking (MeditCom)
Lecture Notes in Computer Science, 2006
This paper describes OntoExtractor a tool for extracting metadata from heterogeneous sources of i... more This paper describes OntoExtractor a tool for extracting metadata from heterogeneous sources of information, producing a “quick-and-dirty” hierarchy of knowledge. This tool is specifically tailored for a quick classification of semi-structured data. By this feature, OntoExtractor is convenient for dealing with a web-based data source.
The changing research culture poses new requirements on research communities across disciplines. ... more The changing research culture poses new requirements on research communities across disciplines. The availability of different research outputs requires appropriate infrastructure in order to publish and recognize them as research contributions. Institutional repositories (IR) provide such support for research outputs generated as part of the research. In this work, we conduct a survey to assess the research data management activities and services for an IR at a university setting. The survey results show an increasing presence of research data (RD) as well as a need for an IR component to support it for the participating communities. The survey reveals that common practices among researchers are to save data into spreadsheets, text documents, and relational databases, which they manage personally without any institutional support. Most of them store this data on a personal computer or portable storage drives. Despite these storing practices, which are not optimal for sharing, most participants have a need of sharing their data, and they do it mostly using email attachments. The survey also reveals features that repositories should provide, mostly RD creation-and dissemination-related activities, but also covering those that relate to backup and access-granting.
HAL (Le Centre pour la Communication Scientifique Directe), 2006
ABSTRACT BRIDGE is a research project which aims at dealing with the problems of e-recruitment by... more ABSTRACT BRIDGE is a research project which aims at dealing with the problems of e-recruitment by considering a new approach based on competency management. The idea consists in allowing a job seeker (respectively a recruiter and a training provider) to identify and formally represent the competencies underlying its resume or Curriculum Vitae (respectively its job offer and its training programme). These competencies, which allow knowledge, skills, abilities, traits and motives acquired by a person (respectively required for a job vacancy and provided by a training programme) to be made explicit, are then used to refine the matching process between "supply and demand". From a technical viewpoint, these competencies correspond to ontology-based annotations, which are formally represented by using Semantic Web languages, in particular OWL (Ontology Web Language) and RDF (Resource Description Framework). By means of the competency dimension and the competency gap measurement, BRIDGE web services will enhance employment market and promote lifelong learning by enabling pertinent access to learning possibilities including eLearning solutions. BRIDGE will also contribute to the enrichment of individual human capital, the preservation of social coherence and the improvement of companies' competitiveness. Indeed, thanks to BRIDGE, users will be able to more efficiently identify lacking competencies between their CV and a list of possible job offers. BRIDGE completes this service by providing information regarding available courses to reduce this lack and to acquire missing competencies.
CommOn (competency management through ontologies) is a framework which aims at developing operati... more CommOn (competency management through ontologies) is a framework which aims at developing operational knowledge-based systems dedicated to the management of competencies. Such a competency-based system can be used for different purposes such as staff development and deployment, job analysis or economic evaluation. CommOn is based on two models (implemented with specific tools) which guide (i) the building of competency reference
Linked Open Data (LOD) and semantic technologies present an enormous potential for a variety of a... more Linked Open Data (LOD) and semantic technologies present an enormous potential for a variety of applications. LOD cloud is changing shape in a permanent way with the continuous addition of new datasets. This raises new possibilities for usage and affects the creation of applications on top of these data for various purposes. The purpose of this paper is to describe the use of LOD and semantic technologies to allow content and search results enrichment for digital libraries, with an emphasis to EconBiz case study. The main challenge for achieving this, present the process of contextualization of scientific publications with semantically related data. More specifically we describe the process of creating a profile of publication from the content that is available in a semantic web representation. This would allow detection and alignment of similar publications from different areas, author details, co-authors relativeness, events, organizations, etc. This is claimed to affect significantly the perspective of what should be of digital libraries in the future.
Proceedings of the 18th International Conference on e-Society (ES 2020), 2020
The changing research culture poses new requirements on research communities across disciplines. ... more The changing research culture poses new requirements on research communities across disciplines. The availability of different research outputs requires appropriate infrastructure in order to publish and recognize them as research contributions. Institutional repositories (IR) provide such support for research outputs generated as part of the research. In this work, we conduct a survey to assess the research data management activities and services for an IR at a university setting. The survey results show an increasing presence of research data (RD) as well as a need for an IR component to support it for the participating communities. The survey reveals that common practices among researchers are to save data into spreadsheets, text documents, and relational databases, which they manage personally without any institutional support. Most of them store this data on a personal computer or portable storage drives. Despite these storing practices, which are not optimal for sharing, most participants have a need of sharing their data, and they do it mostly using email attachments. The survey also reveals features that repositories should provide, mostly RD creation-and dissemination-related activities, but also covering those that relate to backup and access-granting.
Enriching scientific publications with semantically related data
2018 26th Signal Processing and Communications Applications Conference (SIU), 2018
In recent years, the machine learning algorithms commenced to be used widely in signal classifica... more In recent years, the machine learning algorithms commenced to be used widely in signal classification area as well as many other areas. Ensemble learning has become one of the most popular Machine Learning approaches due to the high classification performance it provides. In this study, the application of four fundamental ensemble learning methods (Bagging, Boosting, Stacking, and Voting) with five different classification algorithms (Neural Network, Support Vector Machines, k-Nearest Neighbor, Naive Bayes, and C4.5) with the most optimal parameter values on signal datasets is presented. In the experimental studies, ensemble learning methods were applied on 14 different signal datasets and the results were compared in terms of classification accuracy rates. According to the results, the best classification performance was obtained with the Random Forest algorithm which is a Bagging based method.
Proceedings of the 14th International Conference on Knowledge Technologies and Data-driven Business - i-KNOW '14, 2014
Standard-Nutzungsbedingungen: Dieses Dokument darf zu eigenen wissenschaftlichen Zwecken und zum ... more Standard-Nutzungsbedingungen: Dieses Dokument darf zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen dieses Dokument nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, aufführen, vertreiben oder anderweitig nutzen. Sofern für das Dokument eine Open-Content-Lizenz verwendet wurde, so gelten abweichend von diesen Nutzungsbedingungen die in der Lizenz gewährten Nutzungsrechte. Terms of use: This document may be saved and copied for your personal and scholarly purposes. You are not to copy it for public or commercial purposes, to exhibit the document in public, to perform, distribute or otherwise use the document in public. If the document is made available under a Creative Commons Licence you may exercise further usage rights as specified in the licence.
Linked Open Data (LOD) and semantic technologies present an enormous potential for a variety of a... more Linked Open Data (LOD) and semantic technologies present an enormous potential for a variety of applications. LOD cloud is changing shape in a permanent way with the continuous addition of new datasets. This raises new possibilities for usage and affects the creation of applications on top of these data for various purposes. The purpose of this paper is to describe the use of LOD and semantic technologies to allow content and search results enrichment for digital libraries, with an emphasis to EconBiz case study. The main challenge for achieving this, present the process of contextualization of scientific publications with semantically related data. More specifically we describe the process of creating a profile of publication from the content that is available in a semantic web representation. This would allow detection and alignment of similar publications from different areas, author details, co-authors relativeness, events, organizations, etc. This is claimed to affect significa...
Cette these est consacree, d'une part, a l'etude d'approches d'extraction de cara... more Cette these est consacree, d'une part, a l'etude d'approches d'extraction de caracteristiques structurelles et statistiques et leur fusion dans les systemes d'apprentissage connexionnistes, et d'autre part, a la combinaison de classificateurs et la fusion de leurs decisions. L'etude se focalise essentiellement sur deux problemes de la reconnaissance des formes (rdf) : la parametrisation qui a pour objectif d'extraire et de selectionner l'information pertinente de la forme pour fournir une description aussi compacte et informative que possible, et la discrimination qui consiste essentiellement a classer des formes en vue de la reconnaissance. Pour ces deux directions de recherche nous proposons un ensemble de contributions qui commence par une classification des techniques de selection et d'extraction de caracteristiques en rdf. Nous proposons une nouvelle methode d'extraction de caracteristiques structurelles ainsi qu'une methodologie ...
International Journal of Informatics and Communication Technology (IJ-ICT)
Data mining is a computational approach aiming to discover hidden and valuable information in lar... more Data mining is a computational approach aiming to discover hidden and valuable information in large datasets. It has gained importance recently in the wide area of computational among which many in the domain of Business Informatics. This paper focuses on applications of data mining in Customer Relationship Management (CRM). The core of our application is a classifier based on the naive Bayesian classification. The accuracy rate of the model is determined by doing cross validation. The results demonstrated the applicability and effectiveness of the proposed model. Naive Bayesian classifier reported high accuracy. So the classification rules can be used to support decision making in CRM field. The aim of this study is to apply the data mining model to the banking sector as example case study. This work also contains an example data set related with customers to predict if the client will subscribe a term deposit. The results of the implementation are available on a mobile platform.
The publications stored in Digital Libraries (DL), while authoritative and of high quality, due t... more The publications stored in Digital Libraries (DL), while authoritative and of high quality, due to DL's access model, have limited chances of being linked up with other, up-to-date, resources on the Web. On the other hand, the content amassing in Web 2.0-driven resources in the form of scientific blogs and wikis is continually on the rise. Although with no formal publication " check points " or review committees, their key feature – the up-to-datedness – both in the form of ease of publication, and the opportunity for feedback from the community that follows these resources – is something that Digital Libraries obviously lack. With this proposal, by using semantic web and Linked Data technologies, we want to develop a service that will enrich the user search experience by combining the content from DLs with that of Web 2.0-driven scientific blogs and wikis.
BRIDGE is a research project which aims at dealing with the problems of e-recruitment by consider... more BRIDGE is a research project which aims at dealing with the problems of e-recruitment by considering a new approach based on competency management. The idea consists in allowing a job seeker (respectively a recruiter and a training provider) to identify and formally represent the competencies underlying its resume or Curriculum Vitae (respectively its job offer and its training programme). These competencies, which allow knowledge, skills, abilities, traits and motives acquired by a person (respectively required for a job vacancy and provided by a training programme) to be made explicit, are then used to refine the matching process between "supply and demand". From a technical viewpoint, these competencies correspond to ontology-based annotations, which are formally represented by using Semantic Web languages, in particular OWL (Ontology Web Language) and RDF (Resource Description Framework). By means of the competency dimension and the competency gap measurement, BRIDGE w...
Lecture Notes in Computer Science, 2015
Linked Open Data (LOD) and semantic technologies present an enormous potential for a variety of a... more Linked Open Data (LOD) and semantic technologies present an enormous potential for a variety of applications. LOD cloud is changing shape in a permanent way with the continuous addition of new datasets. This raises new possibilities for usage and affects the creation of applications on top of these data for various purposes. The purpose of this paper is to describe the use of LOD and semantic technologies to allow content and search results enrichment for digital libraries, with an emphasis to EconBiz case study. The main challenge for achieving this, present the process of contextualization of scientific publications with semantically related data. More specifically we describe the process of creating a profile of publication from the content that is available in a semantic web representation. This would allow detection and alignment of similar publications from different areas, author details, co-authors relativeness, events, organizations, etc. This is claimed to affect significantly the perspective of what should be of digital libraries in the future.
Content enrichment of publications stored in different cross domain Digital Libraries can facilit... more Content enrichment of publications stored in different cross domain Digital Libraries can facilitate the scholarly communication in big way. However, current DL still entails limitation of interoperability between cross domain repositories. This paper emphasizes on this limitation and proposes an innovative approach for finding and recommending scientific publications which are stored in disparate repositories. At first Linked Open Data is considered by exploring existing alignments between Econstor and other datasets within the current LOD cloud through the STW Thesaurus. Moreover, incorporation of other relevant metadata is proposed by implementing a data mining approach which improves the semantic relativeness of the publications from the recommended list.
Uploads
Papers by Vladimir Radevski