Suchergebnisse

Intro -- Preface -- Organizing Committee -- Conference Chair -- Organizing Chairs -- Technical Chairs -- Technical Co-Chairs -- International Advisory Committee -- Advisory Committee -- Publication Chair -- Editorial Board -- Organizing Committee -- Technical Committee -- Web and Publicity Committee -- Contents -- About the Editors -- 1 Efficient Video Indexing and Retrieval Using Hierarchical Clustering Technique -- Abstract -- 1 Introduction -- 1.1 Information Retrieval Based on the Content -- 2 Existing System -- 3 Proposed System -- 3.1 Advantage of Proposed Technique -- 3.2 Proposed Architecture -- 3.3 Proposed Algorithm -- 4 Experimental Setup -- 4.1 Image Elements Extraction -- 4.1.1 Pseudocode for Color Histogram Calculation -- 4.2 Indexing of Images -- 4.3 Image Retrieval Using Clustering Technique -- 4.3.1 Psecudocode for Cluster Calculation -- 5 Experimental Outcomes -- 6 Conclusion -- References -- 2 A Data Perturbation Method to Preserve Privacy Using Fuzzy Rules -- Abstract -- 1 Introduction -- 1.1 Privacy Preserving Data Mining -- 1.2 Data Perturbation in PPDM -- 2 Background and Related Work -- 3 Data Perturbation Using Fuzzy Logic -- 3.1 Fuzzy Rule Based Systems -- 4 Proposed Work -- 4.1 Proposed Algorithm -- 5 Results -- 6 Conclusion -- References -- 3 An Enhanced Unsupervised Learning Approach for Sentiment Analysis Using Extraction of Tri-Co-Occurrence Words Phrases -- Abstract -- 1 Introduction -- 2 Associated Work -- 3 Proposed Model -- 3.1 Steps for Preprocessing Dataset -- 3.2 Extract Phrase Patterns -- 3.3 Computation of Semantic Orientation -- 3.4 Pseudocode Procedure for Implementation -- 4 Results and Discussions -- 5 Conclusion and Future Scope -- Acknowledgements -- References -- 4 Land Use/Land Cover Segmentation of Satellite Imagery to Estimate the Utilization of Earth's Surface -- Abstract -- 1 Introduction.

"This book identifies the economic as well as financial problems that may be solved efficiently with computational methods and explains why those problems should best be solved with computational methods" - Provided by publisher

"Computational Intelligence in Urban Infrastructure consolidates experiences and research results in computational intelligence and its applications in urban infrastructure. It discusses various techniques and application areas of smart urban infrastructure including topics related to smart city management. Major topics covered include smart home automation, intelligent lighting, smart human care services, intelligent transportation systems, ontologies in urban development domain, and intelligent monitoring, control, and security of critical infrastructure systems supported by case studies. Features: Covers application of AI and computational intelligence techniques in urban infrastructure planning. Discusses characteristics and features of smart urban management. Explores relationship between smart home and smart city management. Deliberates various smart home techniques. Includes different case studies for supporting and analysing various aspects of smart urban infrastructure management. This book is aimed at researchers, graduate students, libraries in communication networks, urban and town planning as well as civil engineering"--

Computational intelligence is a very active and fruitful research of artificial intelligence with a broad spectrum of applications. Remote sensing data has been a salient field of application of computational intelligence algorithms, both for the exploitation of the data and for the research/ development of new data analysis tools. In this editorial paper we provide the setting of the special issue "Computational Intelligence in Remote Sensing" and an overview of the published papers. The 11 accepted and published papers cover a wide spectrum of applications and computational tools that we try to summarize and put in perspective in this editorial paper. ; FEDER funds for the MINECO project TIN2017-85827-P project CybSPEED 777720 Elkartek 2018 funding program of the Basque Government KK-2018/00071

"This book deals with the computational intelligence field, particularly business applications adopting computational intelligence techniques"--Provided by publisher

Computational intelligence (CI), as an alternative to statistical and econometric approaches, has been applied to a wide range of economics and finance problems in recent years, for example to price forecasting and market efficiency. This book contains research ranging from applications in financial markets and business administration to various economics problems. Not only are empirical studies utilizing various CI algorithms presented, but so also are theoretical models based on computational methods. In addition to direct applications of computational intelligence, readers can also observe

This book introduces and presents the newest up-to-date methods, approaches and technologies on how to detect child cyberbullying on social media as well as monitor kids E-learning, monitor games designed and social media activities for kids. On a daily basis, children are exposed to harmful content online. There have been many attempts to resolve this issue by conducting methods based on rating and ranking as well as reviewing comments to show the relevancy of these videos to children; unfortunately, there still remains a lack of supervision on videos dedicated to kids. This book also introduces a new algorithm for content analysis against harmful information for kids. Furthermore, it establishes the goal to track useful information of kids and institutes detection of kids textual aggression through methods of machine and deep learning and natural language processing for a safer space for children on social media and online and to combat problems, such as lack of supervision, cyberbullying, kids exposure to harmful content. This book is beneficial to postgraduate students and researchers' concerns on recent methods and approaches to kids' cybersecurity.

Das ständig wachsende Bedürfnis nach universell verfügbarer Rechen- und Speicherkapazität wird durch die in den letzten Jahren vorangetriebene Entwicklung neuer Architekturen für die vernetzte Interaktion zwischen Nutzern und Anbietern von Rechenressourcen mehr und mehr erfüllt. Dabei ist die Umsetzung einer Infrastruktur zur koordinierten Nutzung global verteilter Rechenressourcen längst in Forschung und Wirtschaft realisiert worden. Diese als Grid-Computing bezeichnete Infrastruktur wird künftig integraler Bestandteil der globalen Ressourcenlandschaft sein, sodass die Ausführung von lokal eingereichten Berechnungsaufgaben nicht mehr ortsgebunden ist, sondern flexibel zwischen unterschiedlichen Ressourcenanbietern migriert werden kann. Bereits heute sind unterschiedliche Nutzergemeinschaften im Rahmen von Community-Grids in virtuellen Organisationen zusammengefasst, die neben einem gemeinsamen Forschungs- oder Anwendungsinteresse auch häufig eine Menge von IT-Ressourcen gemeinsam nutzen. Ziel ist es aber, auf lange Sicht eine Community-übergreifende Kooperation im Sinne einer globalen Grid-Infrastruktur unter Wahrung lokaler Autonomie weiter zu fördern. Dabei bringt die Interaktion mit anderen Communities im Grid sowohl Chancen als auch Herausforderungen mit sich, da durch die Nutzbarkeit global verteilter Ressourcen auch höhere Anforderungen in Bezug auf Berechnungsgeschwindigkeit und Wartezeiten von Seiten der Nutzer gestellt werden. Der Schlüssel für den effizienten Betrieb künftiger Computational Grids liegt daher in der Entwicklung tragfähiger Architekturen und Strategien für das Scheduling, also in der Zuteilung der Jobs zu den Ressourcen. Bisher sind die Methoden für die Verhandlung von Jobübernahmen zwischen Ressourcenanbietern jedoch nur sehr rudimentär entwickelt. In dieser Arbeit werden deshalb dezentrale Schedulingstrategien für Computational Grids entwickelt und unter Einsatz von Methoden der Computational Intelligence realisiert und optimiert. Dabei werden einzelne virtuelle Organisationen als autonome Einheiten betrachtet, die über eine Annahme oder Abgabe sowohl von eigenen als auch von extern eingereichten Jobs entscheiden. Durch die Beachtung einer restriktiven Informationspolitik werden die Autorität und Sicherheit virtueller Organisationen gewahrt und zugleich wird die Skalierbarkeit in größeren Umgebungen durch den dezentralen Aufbau sichergestellt. Zunächst werden verschiedene dezentrale Strategien entwickelt und simulatorisch untersucht. Die Ergebnisse geben dann Aufschlüsse über die Dynamik und Eigenschaften eines derartigen Verbunds. Auf Basis der so gewonnenen Erkenntnisse werden die Mechanismen zur Entscheidungsfindung verfeinert und in einer neu entworfenen modularen Schedulingarchitektur umgesetzt. Mittels evolutionär optimierter Fuzzy-Systeme wird anschließend die Entscheidungsfindung optimiert. Die Interaktion zwischen virtuellen Organisationen wird dann alternativ mittels co-evolutionärer Algorithmen angepasst. Die auf Basis realer Arbeitslastaufzeichnungen durchgeführten Evaluationen zeigen, dass die so erstellten Grid-Schedulingstrategien für alle am Grid teilnehmenden Communities deutlich verkürzte Antwortzeiten für die jeweiligen Nutzergemeinschaften erreichen. Gleichzeitig wird eine große Robustheit der Verfahren sowohl gegenüber veränderlichen Grid-Umgebungen als auch gegenüber verändertem Nutzerverhalten bewiesen. Die Ergebnisse sind als Motivation für die stärkere Community-übergreifende Kooperation im Sinne eines Computational Grid zu sehen, da dies bei Nutzung entsprechend optimierter Verfahren in einer Win-win Situation für alle Teilnehmer resultiert. ; The ever-growing need for universally available computing and storage capacity is more and more satisfied by new architectures for networked interaction between users and providers of computing resources. Today, research and industry have realized an infrastructure for the coordinated use of globally distributed computing resources. The so-called grid computing infrastructure is assumed to be an integral part of the future global resource landscapes. In such an environment, submitted computing jobs are no longer bound locally, but can be flexibly migrated between different resource providers. In the context of community grids, different users are organized into virtual organizations, which typically share---in addition to a joint research---various computing resources. However, cooperation among different virtual organizations at the moment occurs only very rarely as---besides technical issues---this requires more advanced decentralized grid scheduling concepts. In order to fully utilize the capabilities of a federated grid, it is essential to promote a cross-community cooperation in the sense of a global grid infrastructure. At the same time, however, it is most important that the local autonomy is maintained, as no virtual organization would voluntarily cede the control of their local resources. Thus, the interaction among different communities in the grid brings both opportunities and challenges: The newly formed flexibility makes users even more demanding with respect to computing result delivery and wait times. The efficient operation of future computational grids highly depends on the development of viable architectures and strategies for scheduling, i.e. the allocation of jobs to resources and powerful methods for the negotiation of job migrations. In this work, we therefore develop distributed scheduling strategies for computational grids using various methods from computational intelligence. Different virtual organizations are seen as autonomous entities that decide on the acceptance or decline of jobs. Here, jobs can be offered by other schedulers or by the local user communities. The authority and security of virtual organizations will be maintained by following a restrictive information policy that strongly limits the exchange of system state information. The fully decentralized grid structure guarantees that the scheduling concepts are also applicable in large-scale environments. Initially, several decentralized strategies are developed and tested by extensive simulations with real-world workload traces. The results give first insights into the dynamics and characteristics of the assumed grid federations. Based on these findings, the mechanisms for decision-making is refined and implemented in a newly designed modular scheduling architecture. Using an evolutionary fuzzy system, the decision-making and interaction between virtual organizations is realized and further optimized. In a last step, also a coevolutionary algorithm is applied to improve the scheduling decisions. The evaluation based on real workload recordings reveals that it is possible to achieve significantly shorter response times for all respective user communities. At the same time, we demonstrate a strong robustness of the procedures, both to changing grid environments and changed user behavior. The results can be seen as a motivation for the increased cross-community cooperation in terms of a global computational grid, as this results in a win-win situation for all participants.

Proceedings of the First PhD Symposium on Sustainable Ultrascale Computing Systems (NESUS PhD 2016) Timisoara, Romania. February 8-11, 2016. ; In the pharmaceutical industry, a good understanding of the casual relationship between product quality and attributes of formulations is very useful in developing new products, and optimizing manufacturing processes. Feature selection is mandatory due to the abundance of noisy, irrelevant, or misleading features. The selected features will improve the performance of the prediction model and will provide a faster and more cost effective prediction than using all the features. With the big data captured in the pharmaceutical product development practice, computational intelligence (CI) models and machine learning algorithms could potentially be used to identify the process parameters of formulations and manufacturing processes. That needs a deep investigation of roller compaction process parameters of pharmaceutical formulations that affect the ribbons production. In this work, we are using the bio-inspired optimization algorithms for feature selection such as (grey wolf, Bat, flower pollination, social spider, antlion, moth-flame, genetic algorithms, and particle swarm) to predict the different pharmaceutical properties. ; European Cooperation in Science and Technology. COST ; This work was supported by the IPROCOM Marie Curie initial training network, funded through the People Programme (Marie Curie Actions) of the European Union's Seventh Framework Programme FP7/2007-2013/ under REA grant agreement No. 316555. In addition, this work was partially supported by NESUS.