Pub Date : 2019-11-01DOI: 10.1109/BdKCSE48644.2019.9010603
Yuri P. Pavlov
Theoretical formulations deriving from a theorem of Karni-Jaffray under the light of the Savage and von Neumann theory are discussed in the paper. The main purpose is to develop an approximation based methodology for the assessment of subjective probabilities. The basic information is the decision-maker (DM) preferences explicitly expressed in a cardinal way (‘yes’, ‘no’, ‘no preference’). The elicitation procedure uses the Lpτ pseudo-random Sobol’ sequences and the stochastic approximation approach. The dialog DM-computer is modeled numerically and the results are presented.
{"title":"Estimation of Subjective Probabilities through the Prism of Karni-Jaffray Theorem and Stochastic Approximation","authors":"Yuri P. Pavlov","doi":"10.1109/BdKCSE48644.2019.9010603","DOIUrl":"https://doi.org/10.1109/BdKCSE48644.2019.9010603","url":null,"abstract":"Theoretical formulations deriving from a theorem of Karni-Jaffray under the light of the Savage and von Neumann theory are discussed in the paper. The main purpose is to develop an approximation based methodology for the assessment of subjective probabilities. The basic information is the decision-maker (DM) preferences explicitly expressed in a cardinal way (‘yes’, ‘no’, ‘no preference’). The elicitation procedure uses the Lpτ pseudo-random Sobol’ sequences and the stochastic approximation approach. The dialog DM-computer is modeled numerically and the results are presented.","PeriodicalId":206080,"journal":{"name":"2019 Big Data, Knowledge and Control Systems Engineering (BdKCSE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122639668","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/BdKCSE48644.2019.9010661
Zeinab Nazemi Absardi, R. Javidan
Recent development of DNA sequencing technologies has led to a significant increase in genomic data volume. Such a big amount of genome data needs appropriate data storage, data management, and data transfer policies. Compressing genomes can be used for efficient data management. Auto-encoder is a kind of deep neural networks, due to its ability to reduce the dimension of data is suitable for this purpose. In this paper, a new method for genome compression with auto-encoders based on deep neural networks is proposed. It is the first time that an auto-encoder is used to compress the genomes. Experimental results showed that the proposed method can achieve a compression ratio of up to 5 and 92 percent compression accuracy in case of reference-free genome compression. Moreover, after the auto-encoder training stage, the trained network will have a very short compression time which makes it suitable for real-time applications.
{"title":"A Fast Reference-Free Genome Compression Using Deep Neural Networks","authors":"Zeinab Nazemi Absardi, R. Javidan","doi":"10.1109/BdKCSE48644.2019.9010661","DOIUrl":"https://doi.org/10.1109/BdKCSE48644.2019.9010661","url":null,"abstract":"Recent development of DNA sequencing technologies has led to a significant increase in genomic data volume. Such a big amount of genome data needs appropriate data storage, data management, and data transfer policies. Compressing genomes can be used for efficient data management. Auto-encoder is a kind of deep neural networks, due to its ability to reduce the dimension of data is suitable for this purpose. In this paper, a new method for genome compression with auto-encoders based on deep neural networks is proposed. It is the first time that an auto-encoder is used to compress the genomes. Experimental results showed that the proposed method can achieve a compression ratio of up to 5 and 92 percent compression accuracy in case of reference-free genome compression. Moreover, after the auto-encoder training stage, the trained network will have a very short compression time which makes it suitable for real-time applications.","PeriodicalId":206080,"journal":{"name":"2019 Big Data, Knowledge and Control Systems Engineering (BdKCSE)","volume":"109 6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128701557","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/BdKCSE48644.2019.9010614
Agil Yolchuyev
With the increasing demand for cloud computing, the cloud storage systems are becoming more attractive to companies for their information processing, because of their scalability and low cost. Moreover, for companies having multiple data centers in the different regions to handle and to access big data objects is one of the major problems, (as far as uploading and downloading to/from remote storages are concerned). One of the proposed solutions to this problem is distributed storage: i.e. to slice large objects into small chunks which are then uploaded to different cloud storages. As will be seen, this problem can be formalized as a constrained combinatorial optimization. In this paper, an optimal uploading strategy is developed to meet the various reliability criteria by solving the underlying combinatorial optimization.
{"title":"A Novel Approach for Optimal Data Uploading to the Distributed Cloud Storage Systems","authors":"Agil Yolchuyev","doi":"10.1109/BdKCSE48644.2019.9010614","DOIUrl":"https://doi.org/10.1109/BdKCSE48644.2019.9010614","url":null,"abstract":"With the increasing demand for cloud computing, the cloud storage systems are becoming more attractive to companies for their information processing, because of their scalability and low cost. Moreover, for companies having multiple data centers in the different regions to handle and to access big data objects is one of the major problems, (as far as uploading and downloading to/from remote storages are concerned). One of the proposed solutions to this problem is distributed storage: i.e. to slice large objects into small chunks which are then uploaded to different cloud storages. As will be seen, this problem can be formalized as a constrained combinatorial optimization. In this paper, an optimal uploading strategy is developed to meet the various reliability criteria by solving the underlying combinatorial optimization.","PeriodicalId":206080,"journal":{"name":"2019 Big Data, Knowledge and Control Systems Engineering (BdKCSE)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126816118","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/BdKCSE48644.2019.9010612
B. Vatchova, David Sanders, M. Adda, A. Gegov
The present paper discusses the use of intelligent control for modelling complex interconnected processes. The latter usually have many inputs and outputs and can be found in various areas of application. While part of the inputs are measurable, others are not due to the presence of stochastic environmental factors. For this reason such kind of processes operate under uncertainty. The latter is addressed in this paper by intelligent systems that use probabilistic and fuzzy network structures.
{"title":"Knowledge Based Modelling of Complex Interconnected Systems","authors":"B. Vatchova, David Sanders, M. Adda, A. Gegov","doi":"10.1109/BdKCSE48644.2019.9010612","DOIUrl":"https://doi.org/10.1109/BdKCSE48644.2019.9010612","url":null,"abstract":"The present paper discusses the use of intelligent control for modelling complex interconnected processes. The latter usually have many inputs and outputs and can be found in various areas of application. While part of the inputs are measurable, others are not due to the presence of stochastic environmental factors. For this reason such kind of processes operate under uncertainty. The latter is addressed in this paper by intelligent systems that use probabilistic and fuzzy network structures.","PeriodicalId":206080,"journal":{"name":"2019 Big Data, Knowledge and Control Systems Engineering (BdKCSE)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114703306","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/BdKCSE48644.2019.9010668
Gernot Steindl, W. Kastner
In this paper, a Smart Data Service, based on Semantic Web technology is introduced, which supports the control engineer during the data-driven model development process by enabling enhanced data analysis. As a perquisite for such a service, sensor data consisting of semantic meta data as well as time series data have to be integrated into a so-called knowledge graph. Therefore, three different integration approaches, found in the literature, were evaluated and compared regarding their query execution performance. The characteristics and limitations of these three methods are discussed to specify the conditions for their specific utilization.
{"title":"Query Performance Evaluation of Sensor Data Integration Methods for Knowledge Graphs","authors":"Gernot Steindl, W. Kastner","doi":"10.1109/BdKCSE48644.2019.9010668","DOIUrl":"https://doi.org/10.1109/BdKCSE48644.2019.9010668","url":null,"abstract":"In this paper, a Smart Data Service, based on Semantic Web technology is introduced, which supports the control engineer during the data-driven model development process by enabling enhanced data analysis. As a perquisite for such a service, sensor data consisting of semantic meta data as well as time series data have to be integrated into a so-called knowledge graph. Therefore, three different integration approaches, found in the literature, were evaluated and compared regarding their query execution performance. The characteristics and limitations of these three methods are discussed to specify the conditions for their specific utilization.","PeriodicalId":206080,"journal":{"name":"2019 Big Data, Knowledge and Control Systems Engineering (BdKCSE)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126003379","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/BdKCSE48644.2019.9010593
Noman Shabbir, L. Kütt, M. Jawad, Roya Amadiahanger, M. N. Iqbal, A. Rosin
Wind energy forecasting is a very challenging task as it involves many variable factors from wind speed, weather season, location and many other factors. Its accurate prediction can be quite helpful in maintaining the balance between demand and supply, and issues related to the reliability of a power system. In this article, the Recurrent Neural Network (RNN) based forecasting algorithm is used for the three day-ahead predictions of energy generation from wind sources in Estonia. Then a comparison is made between the predicted energy generation of Estonian energy regulatory authority's algorithm and this RNN based algorithm. The simulation results show that our proposed algorithm has lower Root Mean Square Error (RMSE) value and it gives better forecasting.
{"title":"Wind Energy Forecasting Using Recurrent Neural Networks","authors":"Noman Shabbir, L. Kütt, M. Jawad, Roya Amadiahanger, M. N. Iqbal, A. Rosin","doi":"10.1109/BdKCSE48644.2019.9010593","DOIUrl":"https://doi.org/10.1109/BdKCSE48644.2019.9010593","url":null,"abstract":"Wind energy forecasting is a very challenging task as it involves many variable factors from wind speed, weather season, location and many other factors. Its accurate prediction can be quite helpful in maintaining the balance between demand and supply, and issues related to the reliability of a power system. In this article, the Recurrent Neural Network (RNN) based forecasting algorithm is used for the three day-ahead predictions of energy generation from wind sources in Estonia. Then a comparison is made between the predicted energy generation of Estonian energy regulatory authority's algorithm and this RNN based algorithm. The simulation results show that our proposed algorithm has lower Root Mean Square Error (RMSE) value and it gives better forecasting.","PeriodicalId":206080,"journal":{"name":"2019 Big Data, Knowledge and Control Systems Engineering (BdKCSE)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125055379","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/BdKCSE48644.2019.9010600
V. Terzieva, E. Paunova-Hubenova, Katia Todorova, P. Kademova-Katzarova
Recently, e-learning resources become widespread in school education worldwide, as they are a prerequisite for an efficient learning process. These resources are scattered across many websites and their search takes too long. Despite their diversity, often it is difficult for teachers to find the proper resources. After analyzing the findings of an anonymous online survey of Bulgarian teachers on the use of ICT and e-learning resources, researchers identified the need for easy access to various e-resources for school education. The paper offers a concept for a centralized access portal where users to upload links to e-learning resources that are shortly described and classified according to several indicators. These include accessibility, type, subject, school grade, purpose and other meaningful parameters. Free registration for teachers and students will allow data to be aggregated by user groups, to carry out analyzes of the resources' usage, to draw trends, and to make conclusions for future policies.
{"title":"Learning Analytics - Need of Centralized Portal for Access to E-Learning Resources","authors":"V. Terzieva, E. Paunova-Hubenova, Katia Todorova, P. Kademova-Katzarova","doi":"10.1109/BdKCSE48644.2019.9010600","DOIUrl":"https://doi.org/10.1109/BdKCSE48644.2019.9010600","url":null,"abstract":"Recently, e-learning resources become widespread in school education worldwide, as they are a prerequisite for an efficient learning process. These resources are scattered across many websites and their search takes too long. Despite their diversity, often it is difficult for teachers to find the proper resources. After analyzing the findings of an anonymous online survey of Bulgarian teachers on the use of ICT and e-learning resources, researchers identified the need for easy access to various e-resources for school education. The paper offers a concept for a centralized access portal where users to upload links to e-learning resources that are shortly described and classified according to several indicators. These include accessibility, type, subject, school grade, purpose and other meaningful parameters. Free registration for teachers and students will allow data to be aggregated by user groups, to carry out analyzes of the resources' usage, to draw trends, and to make conclusions for future policies.","PeriodicalId":206080,"journal":{"name":"2019 Big Data, Knowledge and Control Systems Engineering (BdKCSE)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128289158","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/BdKCSE48644.2019.9010597
A. Bakanov, T. Atanasova, N. Bakanova
The paper proposes an approach to modelling human interaction with the distributed intellectual information environment, using the system of “power-society” as an example. The approach involves the consideration of the human factor in modelling of information interaction. The human factor in this article is considered simplified, as a set of cognitive characteristics of the user. Within the framework of the developed approach, it is proposed, in the process of human-computer interaction, to implicitly test users in order to obtain information about the cognitive styles of each specific user, the effectiveness and efficiency of human-computer interaction, and the degree of subjective satisfaction from the human-computer interaction. Since the information system “power-society” will operate on a national scale, the processing of collected data is supposed to be carried out using the technology of “big data”. The proposed approach allows us to take into account the influence of a person's cognitive equalities on indicators of his interaction with the information systems that implement a set of public services on the Internet.
{"title":"Cognitive Approach to Modeling Human-Computer Interaction with a Distributed Intellectual Information Environment","authors":"A. Bakanov, T. Atanasova, N. Bakanova","doi":"10.1109/BdKCSE48644.2019.9010597","DOIUrl":"https://doi.org/10.1109/BdKCSE48644.2019.9010597","url":null,"abstract":"The paper proposes an approach to modelling human interaction with the distributed intellectual information environment, using the system of “power-society” as an example. The approach involves the consideration of the human factor in modelling of information interaction. The human factor in this article is considered simplified, as a set of cognitive characteristics of the user. Within the framework of the developed approach, it is proposed, in the process of human-computer interaction, to implicitly test users in order to obtain information about the cognitive styles of each specific user, the effectiveness and efficiency of human-computer interaction, and the degree of subjective satisfaction from the human-computer interaction. Since the information system “power-society” will operate on a national scale, the processing of collected data is supposed to be carried out using the technology of “big data”. The proposed approach allows us to take into account the influence of a person's cognitive equalities on indicators of his interaction with the information systems that implement a set of public services on the Internet.","PeriodicalId":206080,"journal":{"name":"2019 Big Data, Knowledge and Control Systems Engineering (BdKCSE)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127344570","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/BdKCSE48644.2019.9010591
P. Giannakopoulou, P. Chountas
The objective of this project is twofold. Firstly, researchers wants to identify between multiple financial and shipping related measures, the features that have statistically significant impact on the estimation of the spot voyage time charter price of P1A Panamax shipping route which is daily issued from the London-based Baltic Exchange. Secondly, significant objective of this thesis is to examine the predictive ability of multiple multivariant feature models (based on the results of the first objective) and of single variable time series models answering to the question “What is the estimated voyage time charter price of P1A shipping route for tomorrow?”.
{"title":"Forecasting the Spot Price of P1A Shipping Route","authors":"P. Giannakopoulou, P. Chountas","doi":"10.1109/BdKCSE48644.2019.9010591","DOIUrl":"https://doi.org/10.1109/BdKCSE48644.2019.9010591","url":null,"abstract":"The objective of this project is twofold. Firstly, researchers wants to identify between multiple financial and shipping related measures, the features that have statistically significant impact on the estimation of the spot voyage time charter price of P1A Panamax shipping route which is daily issued from the London-based Baltic Exchange. Secondly, significant objective of this thesis is to examine the predictive ability of multiple multivariant feature models (based on the results of the first objective) and of single variable time series models answering to the question “What is the estimated voyage time charter price of P1A shipping route for tomorrow?”.","PeriodicalId":206080,"journal":{"name":"2019 Big Data, Knowledge and Control Systems Engineering (BdKCSE)","volume":"39 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113933939","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/BdKCSE48644.2019.9010596
Ivan Gaidarski, P. Kutinchev
The paper present our approach for protecting sensitive data, using the methods of Big Data. To effectively protect the valuable information within the organization, the following steps are needed: Employing a holistic approach for data classification, identifying sensitive data of the organization, Identifying critical exit points - communication channels, applications and connected devices and protecting the sensitive data by controlling the critical exit points. Our approach is based on creating of component-based architecture framework for ISS, conceptual models for data protection and implementation with COTS IT security products as Data Leak Prevention (DLP) solutions. Our approach is data centric, which is holistic by its nature to protect the meaningful data of the organization.
{"title":"Using Big Data for Data Leak Prevention","authors":"Ivan Gaidarski, P. Kutinchev","doi":"10.1109/BdKCSE48644.2019.9010596","DOIUrl":"https://doi.org/10.1109/BdKCSE48644.2019.9010596","url":null,"abstract":"The paper present our approach for protecting sensitive data, using the methods of Big Data. To effectively protect the valuable information within the organization, the following steps are needed: Employing a holistic approach for data classification, identifying sensitive data of the organization, Identifying critical exit points - communication channels, applications and connected devices and protecting the sensitive data by controlling the critical exit points. Our approach is based on creating of component-based architecture framework for ISS, conceptual models for data protection and implementation with COTS IT security products as Data Leak Prevention (DLP) solutions. Our approach is data centric, which is holistic by its nature to protect the meaningful data of the organization.","PeriodicalId":206080,"journal":{"name":"2019 Big Data, Knowledge and Control Systems Engineering (BdKCSE)","volume":"254 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115629990","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：