Studies in Computational Intelligence 648 Svetozar Margenov Galia Angelova Gennady Agre Editors Innovative Approaches and Solutions in Advanced Intelligent Systems Studies in Computational Intelligence Volume 648 Series editor Janusz Kacprzyk, Polish Academy of Sciences, Warsaw, Poland e-mail: [email protected] About this Series The series “Studies in Computational Intelligence” (SCI) publishes new develop- mentsandadvancesinthevariousareasofcomputationalintelligence—quicklyand with a high quality. The intent is to cover the theory, applications, and design methods of computational intelligence, as embedded in the fields of engineering, computer science, physics and life sciences, as well as the methodologies behind them. The series contains monographs, lecture notes and edited volumes in computational intelligence spanning the areas of neural networks, connectionist systems, genetic algorithms, evolutionary computation, artificial intelligence, cellular automata, self-organizing systems, soft computing, fuzzy systems, and hybrid intelligent systems. Of particular value to both the contributors and the readership are the short publication timeframe and the worldwide distribution, which enable both wide and rapid dissemination of research output. More information about this series at http://www.springer.com/series/7092 Svetozar Margenov Galia Angelova (cid:129) Gennady Agre Editors Innovative Approaches and Solutions in Advanced Intelligent Systems 123 Editors Svetozar Margenov GennadyAgre InstituteofInformationandCommunication InstituteofInformationandCommunication Technologies Technologies BulgarianAcademy of Sciences BulgarianAcademy of Sciences Sofia Sofia Bulgaria Bulgaria GaliaAngelova InstituteofInformationandCommunication Technologies BulgarianAcademy of Sciences Sofia Bulgaria ISSN 1860-949X ISSN 1860-9503 (electronic) Studies in Computational Intelligence ISBN978-3-319-32206-3 ISBN978-3-319-32207-0 (eBook) DOI 10.1007/978-3-319-32207-0 LibraryofCongressControlNumber:2016936578 ©SpringerInternationalPublishingSwitzerland2016 Thisworkissubjecttocopyright.AllrightsarereservedbythePublisher,whetherthewholeorpart of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission orinformationstorageandretrieval,electronicadaptation,computersoftware,orbysimilarordissimilar methodologynowknownorhereafterdeveloped. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publicationdoesnotimply,evenintheabsenceofaspecificstatement,thatsuchnamesareexemptfrom therelevantprotectivelawsandregulationsandthereforefreeforgeneraluse. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authorsortheeditorsgiveawarranty,expressorimplied,withrespecttothematerialcontainedhereinor foranyerrorsoromissionsthatmayhavebeenmade. Printedonacid-freepaper ThisSpringerimprintispublishedbySpringerNature TheregisteredcompanyisSpringerInternationalPublishingAGSwitzerland Preface This volume is a selected collection of papers presented and discussed at the International Conference “Advanced Computing for Innovation (AComIn 2015)”. The papers report innovative approaches and solutions in hot topics of computa- tional intelligence—advanced computing, language and semantic technologies, signal and image processing, as well as optimisation and intelligent control. Advanced Computing is presented by five papers. The work of I. Dimov and V. Todorov is related to efficient methods and tools for analysis of reliability of large-scale models. The authors present an error analysis of an almost optimal MonteCarloalgorithmbasedonbalancingthesystematicandstochasticerror.This contribution to the solution of hard computational problems is relevant to envi- ronmental sciences and computational physics. Three other papers concern high-performance computing in engineering and environmental problems. S. Stoykov and S. Margenov propose and analyse some numerical methods for computingnonlinearfrequency-responsecurvesofplateswithcomplexgeometries. They show that parametric study on the dynamics of complex structures can be carriedoutdeployingappropriateparallelimplementation.K.Lioliosetal.describe a numerical simulation of biochemical oxygen demand removal in Horizontal Subsurface Flow Constructed Wetlands (HSF CW). The simulated experimental data is obtained from five pilot-scale HSF CW units. A. Liolios et al. present an approach that can be effectively used for numerical investigation the seismic inelastic behaviour of cultural heritage industrial buildings of reinforced concrete strengthenedbycableelementsandsubjectedtomultipleearthquakes.Theauthors propose an approach for selecting the optimal cable-bracing scheme by using computeddamageindices.The paper ofJ. Kohler etal.isrelatedtoapplicationof cloud computing for processing Big Data. The authors propose an original query rewriting approach that parallelises queries and joins in an order that has been implemented and tested for performance gains. The approach can be used for improving data security and privacy, especially in public cloud computing environments. v vi Preface Five papers present innovative approaches in the areas of Language, Semantic and Content Technologies. The paper of V. Cantoni et al. is focused on digital preservation of cultural heritage for research and education. It describes an innovative use of interactive digital technologies in cultural heritage presentation practices exemplified by multimodal interaction modalities developed for the Exhibition “1525–2015. Pavia, the Battle, the Future. Nothing was the same again”—a satellite event of the Universal Exposition Expo 2015 in Milan. Two papers are devoted to application of data mining methods for analysing Big and Educational Data. S. Boytcheva et al. present a novel cascade data mining approach for frequent pattern mining, sequence mining and periodical events mining applied to discovery of complex temporal relationships between disorders and their treatment. An evaluation of the approach on real data is provided. I. Nikolova et al. describe application of educational data mining to a real educa- tional web portal—UCHA.SE with the goal to improve the quality of the educa- tional services provided by the site and its revenue generation. The extracted predictive rules are used to make recommendations to the UCHA.SE development team.Twootherpapersarerelatedtoadvancedmethodsandtoolsforprocessingof textual and semantic data. O. Kanishcheva and G. Angelova present an original integrated approach for word sense disambiguation of image tags that can be appliedforimprovingmachinetranslationoftagsorimagesimilaritymeasurement. K. Simov et al. describe an approach for the enrichment of word sense disam- biguation knowledge bases with data-driven relations from a gold standard corpus (annotated with word senses, valency information, syntactic analyses, etc.). The paperisfocusedonBulgarianandEnglishasusecases,buttheapproachisscalable to other languages as well. SignalandImageProcessingissuesarediscussedineightpapers.TheworkofA. Nikolov et al. is related to advanced methods for biometric analysis. The authors describe a novel multimodal ear database characterised by different types of ear representation, either 2D or 3D, depending on the device used for data acquisition. Thedatabasecanbeusedasabenchmarktotestdifferentpatternrecognitionmethods on a set of images capturedin known conditions, and tohighlight the strengths and the weaknesses of each approach in terms of recognition accuracy and robustness. Several problems of smart multi-sensor signal and image processing are treated in the following six papers. S. Ilchev and Z. Ilcheva propose a modular digital water- marking service coupled with steganalysis suitable both for commercial and non commercialusers.The service actsasintermediary facilitatingthe payment flow for commercialusesandisabletogatherandprovidestatisticsaboutimagedistribution and popularity. S. Harizanov investigates several techniques for restoring images corrupted by a non-invertible or ill-conditioned linear transformations and Poisson noise. These techniques are based on image domain decomposition and give rise to multi-constraintoptimisationproblems.D.Karastoyanovetal.presentanewtypeof graphicalBraillescreencontainingamatrixwithlinearelectromagneticmicrodrives and non-magnetic needles, passing through the axes of the electromagnets. P. Koprinkova-Hristova et al. discuss an application of a recently developed smart approach for feature extraction from multi-dimensional data sets using Echo state Preface vii networks to the focalised spectra obtained from multi-sensor measurements of an acousticcamera.Theaimofthestudyisdevelopmentofaremotediagnosticsystem forpredictionofbearingwearingout.V.Kudriashovproposesnewdetectionrulesfor multistaticreceptionofnon-stationaryrandomWieneracousticsignals.Therulesare suitablefor suchapplicationsasmonitoringaircraftenginenoise atlanding/takeoff, and testing of a car engine. I. Chirka suggests two novel techniques for the inter- polation of acoustic fields generated by a single acoustic source based on sinewave modelandinstantaneousphasemeasurement,andgenerationofvirtualmicrophones. The approaches allow improving the accuracy of source localisation avoiding the necessity of using expensive equipment. Some speech analysis problems are con- sideredinthepaperofP.MitankinandS.Mihov,whoproposeanoriginalalgorithm to be integrated in the decoding stage of the speech recognition pipeline. The innovative aspect is the implicit generation of a much larger set of decoding candidates than the state-of-the-art N-best approach. Three papers present innovative results in the areas of Optimisation and Intelligent Control of Traffic. S. Fidanova proposes a model of the passenger flow suitableforanalysingexistingpublictransportationsystems.Thetaskisdefinedasa multi-objective optimisation problem that can be solved by an ant colony optimi- sation algorithm. T. Stoilov et al. introduce an innovative idea for the formal descriptionoftheurbantrafficcontrolbasedonabi-levelmodel.Suchanapproach gives the potential for increasing the space of the control allowing simultaneously the minimisation of the waiting vehicles and maximisation of the traffic flows. A. Balabanov proposes an original algorithm based on fast finding of solutions of Riccati equations for synthesis of a steady-state Kalman estimator, which can be used for onlineapplications such asestimations of thereal traffic intensity. We would like to thank the members of the ACOMIN 2015 Programme Committee,whoreviewedthesubmissionsthoroughlyandfairly,andtheAComIn project (Advanced Computing for Innovation, FP7 Capacity grant 316087) for the generous support provided to the ACOMIN 2015 conference. February 2016 Svetozar Margenov Galia Angelova Gennady Agre Organization Committee The International Conference Advanced Computing for Innovation (ACOMIN 2015) was organised by the Institute of Information and Communication TechnologiesattheBulgarianAcademyofSciences(IICT-BAS)intheframeofthe FP7 project “Advanced Computing for Innovation” (AComIn) grant agreement 316087.Theconferencewasheldduring10–11November2015inSofia,Bulgaria and was aimed at providing a forum for international scientific exchange between Central/Eastern Europe and the rest of the world in several fundamental for com- putational intelligence topics enabling radical progress and development of novel applications. Conference Co-chairs Svetozar Margenov, IICT-BAS Galia Angelova, IICT-BAS Chair of the Organizing Committee Gennady Agre, IICT-BAS Program Committee Hassane Abouaïssa, Université Lille Nord, France Gennady Agre, IICT-BAS, Bulgaria Kiril Alexiev, IICT-BAS, Bulgaria Asen Asenov, University of Glasgow, UK Christopher M. Bishop, Microsoft Research, Cambridge, UK ix x OrganizationCommittee Diego Calvanese, Free University of Bozen-Bolzano, Italy Virginio Cantoni, University of Pavia, Italy Darina Dicheva, Winston-Salem State University, USA Dimo Dimov, IICT-BAS, Bulgaria Ivan Dimov, IICT-BAS, Bulgaria Milena Dobreva, University of Malta, Malta John Domingue, Knowledge Media Institute, The Open University, UK Michel Fliess, ParisTech, LIX—École Polytechnique, France Krasimir Georgiev, IICT-BAS, Bulgaria Oleg Iliev, Fraunhofer Institute for Industrial Mathematics ITWM, Germany Cedric Join, University Laurraine, France Ivan Kalaykov, Örebro University, Sweden Dimitar Karastoyanov, IICT-BAS, Bulgaria Johannes Kraus, University of Duisburg-Essen, Germany Sergei O. Kuznetsov, National Research University Higher School of Economics, Moscow, Russia Yann LeCun, New York University, USA Markos Papageorgiou, Technical University of Crete, Greece Ioannis Papamichail, Technical University of Crete, Greece Vincenzo Piuri, University of Milan, Italy Siegfried Selberherr, The Vienna University of Technology Kiril Simov, IICT-BAS, Bulgaria Todor Stoilov, IICT-BAS, Bulgaria Vidar Thomee, Chalmers University of Technology, Göteborg, Sweden