Pub Date : 2019-01-01DOI: 10.14311/nnw.2019.29.016
Haydar Kiliç, Uğur Yüzgeç, C. Karakuzu
Antlion optimizer algorithm (ALO) is inspired by hunting strategy of antlions. In this study, an improved antlion optimization algorithm is proposed for training parameters of adaptive neuro fuzzy inference system (ANFIS). In the standard ALO algorithm, the greatest deficiency is its long running time during optimization process. The random walking model of ants, the selection procedure and boundary checking mechanism have been developed to speed up standard ALO algorithm. To evaluate the performance of the improved antlion optimization algorithm (IALO), it has been tested on dynamic system modelling problems. ANFIS’s parameters has been optimized by IALO algorithm to model five dynamic systems. ANFIS training procedure has been performed with 30 independent runs. Each training has been started with the random initial parameters of ANFIS and performance metrics have been obtained at the end of training. The results show that the IALO algorithm is able to provide competitive results in terms of mean, best, worst, standard deviation, training time metrics. According to the training time result, the proposed IALO algorithm has better performance than standard ALO algorithm and the average training time has been reduced to approximately 80 %.
{"title":"IMPROVED ANTLION OPTIMIZER ALGORITHM AND ITS PERFORMANCE ON NEURO FUZZY INFERENCE SYSTEM","authors":"Haydar Kiliç, Uğur Yüzgeç, C. Karakuzu","doi":"10.14311/nnw.2019.29.016","DOIUrl":"https://doi.org/10.14311/nnw.2019.29.016","url":null,"abstract":"Antlion optimizer algorithm (ALO) is inspired by hunting strategy of antlions. In this study, an improved antlion optimization algorithm is proposed for training parameters of adaptive neuro fuzzy inference system (ANFIS). In the standard ALO algorithm, the greatest deficiency is its long running time during optimization process. The random walking model of ants, the selection procedure and boundary checking mechanism have been developed to speed up standard ALO algorithm. To evaluate the performance of the improved antlion optimization algorithm (IALO), it has been tested on dynamic system modelling problems. ANFIS’s parameters has been optimized by IALO algorithm to model five dynamic systems. ANFIS training procedure has been performed with 30 independent runs. Each training has been started with the random initial parameters of ANFIS and performance metrics have been obtained at the end of training. The results show that the IALO algorithm is able to provide competitive results in terms of mean, best, worst, standard deviation, training time metrics. According to the training time result, the proposed IALO algorithm has better performance than standard ALO algorithm and the average training time has been reduced to approximately 80 %.","PeriodicalId":49765,"journal":{"name":"Neural Network World","volume":"1 1","pages":""},"PeriodicalIF":0.8,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67123346","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-01-01DOI: 10.14311/NNW.2019.29.008
A. Ata, Muhammad Adnan Khan, Sagheer Abbas, Gulzar Ahmad, A. Fatima
: By the dramatic growth of the population in cities requires the traf-fic systems to be designed efficiently and sustainably by taking full advantage of modern-day technology. Dynamic traffic flow is a significant issue which brings about a block of traffic movement. Thus, for tackling this issue, this paper aims to provide a mechanism to predict the traffic congestion with the help of Artificial Neural Networks (ANN) which shall control or minimize the blockage and result in the smoothening of road traffic. Proposed Modeling Smart Road Traffic Congestion Control using Artificial Back Propagation Neural Networks (MSR2C-ABPNN) for road traffic increase transparency, availability and efficiency in services offered to the citizens. In this paper, the prediction of congestion is operationalized by using the algorithm of backpropagation to train the neural network. The proposed system aims to provide a solution that will increase the comfort level of travellers to make intelligent and better transportation decision, and the neural network is a plausible approach to find traffic situations. Proposed MSR2C-ABPNN with Time series gives attractive results concerning MSE as compared to the fitting approach.
{"title":"MODELLING SMART ROAD TRAFFIC CONGESTION CONTROL SYSTEM USING MACHINE LEARNING TECHNIQUES","authors":"A. Ata, Muhammad Adnan Khan, Sagheer Abbas, Gulzar Ahmad, A. Fatima","doi":"10.14311/NNW.2019.29.008","DOIUrl":"https://doi.org/10.14311/NNW.2019.29.008","url":null,"abstract":": By the dramatic growth of the population in cities requires the traf-fic systems to be designed efficiently and sustainably by taking full advantage of modern-day technology. Dynamic traffic flow is a significant issue which brings about a block of traffic movement. Thus, for tackling this issue, this paper aims to provide a mechanism to predict the traffic congestion with the help of Artificial Neural Networks (ANN) which shall control or minimize the blockage and result in the smoothening of road traffic. Proposed Modeling Smart Road Traffic Congestion Control using Artificial Back Propagation Neural Networks (MSR2C-ABPNN) for road traffic increase transparency, availability and efficiency in services offered to the citizens. In this paper, the prediction of congestion is operationalized by using the algorithm of backpropagation to train the neural network. The proposed system aims to provide a solution that will increase the comfort level of travellers to make intelligent and better transportation decision, and the neural network is a plausible approach to find traffic situations. Proposed MSR2C-ABPNN with Time series gives attractive results concerning MSE as compared to the fitting approach.","PeriodicalId":49765,"journal":{"name":"Neural Network World","volume":"126 1","pages":""},"PeriodicalIF":0.8,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67121796","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-01-01DOI: 10.14311/NNW.2019.29.011
D. Fister, Johnathan Mun, Vita Jagrič, Timotej Jagrič
Deep-learning initiatives have vastly changed the analysis of data. Complex networks became accessible to anyone in any research area. In this paper we are proposing a deep-learning long short-term memory network (LSTM) for automated stock trading. A mechanical trading system is used to evaluate its performance. The proposed solution is compared to traditional trading strategies, i.e., passive and rule-based trading strategies, as well as machine learning classifiers. We have discovered that the deep-learning long short-term memory network has outperformed other trading strategies for the German blue-chip stock, BMW, during the 2010–2018 period.
{"title":"DEEP LEARNING FOR STOCK MARKET TRADING: A SUPERIOR TRADING STRATEGY?","authors":"D. Fister, Johnathan Mun, Vita Jagrič, Timotej Jagrič","doi":"10.14311/NNW.2019.29.011","DOIUrl":"https://doi.org/10.14311/NNW.2019.29.011","url":null,"abstract":"Deep-learning initiatives have vastly changed the analysis of data. Complex networks became accessible to anyone in any research area. In this paper we are proposing a deep-learning long short-term memory network (LSTM) for automated stock trading. A mechanical trading system is used to evaluate its performance. The proposed solution is compared to traditional trading strategies, i.e., passive and rule-based trading strategies, as well as machine learning classifiers. We have discovered that the deep-learning long short-term memory network has outperformed other trading strategies for the German blue-chip stock, BMW, during the 2010–2018 period.","PeriodicalId":49765,"journal":{"name":"Neural Network World","volume":"1 1","pages":""},"PeriodicalIF":0.8,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67123148","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-01-01DOI: 10.14311/nnw.2019.29.012
Alena Rybičková, D. Mocková, D. Teichmann
This paper focuses on the continuous location-routing problem that comprises of the location of multiple depots from a given region and determining the routes of vehicles assigned to these depots. The objective of the problem is to design the delivery system of depots and routes so that the total cost is minimal. The standard location-routing problem considers a finite number of possible locations. The continuous location-routing problem allows location to infinite number of locations in a given region and makes the problem much more complex. We present a genetic algorithm that tackles both location and routing subproblems simultaneously.
{"title":"GENETIC ALGORITHM FOR THE CONTINUOUS LOCATION-ROUTING PROBLEM","authors":"Alena Rybičková, D. Mocková, D. Teichmann","doi":"10.14311/nnw.2019.29.012","DOIUrl":"https://doi.org/10.14311/nnw.2019.29.012","url":null,"abstract":"This paper focuses on the continuous location-routing problem that comprises of the location of multiple depots from a given region and determining the routes of vehicles assigned to these depots. The objective of the problem is to design the delivery system of depots and routes so that the total cost is minimal. The standard location-routing problem considers a finite number of possible locations. The continuous location-routing problem allows location to infinite number of locations in a given region and makes the problem much more complex. We present a genetic algorithm that tackles both location and routing subproblems simultaneously.","PeriodicalId":49765,"journal":{"name":"Neural Network World","volume":"1 1","pages":""},"PeriodicalIF":0.8,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67123189","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-01-01DOI: 10.14311/nnw.2019.29.023
P. Bouchner, M. Novák, Z. Votruba
Artificial systems play an extremely important role in human life. Each day, almost all people on the Earth have to interact with various complex systems, which are of a very different nature and target application. These all system structures and their whole sets can be of various degrees of complexity and can be discriminated into many categories. These three can be considered as their main kinds:
{"title":"Editorial: How can artificial systems rise in a tool for mind?","authors":"P. Bouchner, M. Novák, Z. Votruba","doi":"10.14311/nnw.2019.29.023","DOIUrl":"https://doi.org/10.14311/nnw.2019.29.023","url":null,"abstract":"Artificial systems play an extremely important role in human life. Each day, almost all people on the Earth have to interact with various complex systems, which are of a very different nature and target application. These all system structures and their whole sets can be of various degrees of complexity and can be discriminated into many categories. These three can be considered as their main kinds:","PeriodicalId":49765,"journal":{"name":"Neural Network World","volume":"1 1","pages":""},"PeriodicalIF":0.8,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67123174","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-01-01DOI: 10.14311/NNW.2019.29.002
Jakub Snor, Jaromir Kukal, Quang Van Tran
The self organization can be performed in an Euclidean space as usually defined or in any metric space which is generalization of previous one. Both approaches have advantages and disadvantages. A novel method of batch SOM learning is designed to yield from the properties of the Hilbert space. This method is able to operate with finite or infinite dimensional patterns from vector space using only their scalar product. The paper is focused on the formulation of objective function and algorithm for its local minimization in a discrete space of partitions. General methodology is demonstrated on pattern sets from a space of functions.
{"title":"SOM IN HILBERT SPACE","authors":"Jakub Snor, Jaromir Kukal, Quang Van Tran","doi":"10.14311/NNW.2019.29.002","DOIUrl":"https://doi.org/10.14311/NNW.2019.29.002","url":null,"abstract":"The self organization can be performed in an Euclidean space as usually defined or in any metric space which is generalization of previous one. Both approaches have advantages and disadvantages. A novel method of batch SOM learning is designed to yield from the properties of the Hilbert space. This method is able to operate with finite or infinite dimensional patterns from vector space using only their scalar product. The paper is focused on the formulation of objective function and algorithm for its local minimization in a discrete space of partitions. General methodology is demonstrated on pattern sets from a space of functions.","PeriodicalId":49765,"journal":{"name":"Neural Network World","volume":"1 1","pages":""},"PeriodicalIF":0.8,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67120199","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-01-01DOI: 10.14311/nnw.2019.29.014
Hiam Alquran, Ali Mohammad Alqudah, Isam Abu-Qasmieh, Alaa Al-Badarneh, S. Almashaqbeh
Electrocardiogram (ECG) is one of the most important and effective tools in clinical routine to assess the cardiac arrhythmias. In this research higherorder spectral estimations, bispectrum and third-order cumulants, are evaluated, saved, and pre-trained using convolutional neural networks (CNN) algorithm. CNN is transferred in this study to carry out automatic ECG arrhythmia diagnostics after employing the higher-order spectral algorithms. Transfer learning strategies are applied on pre-trained convolutional neural network, namely AlexNet and GoogleNet, to carry out the final classification. Five different arrhythmias of ECG waveform are chosen from the MIT-BIH arrhythmia database to evaluate the proposed approach. The main contribution of this study is to utilize the pre-trained convolutional neural networks with a combination of higher-order spectral estimations of arrhythmias ECG signal to implement a reliable and applicable deep learning classification technique. The Highest average accuracy obtained is 97.8 % when using third cumulants and GoogleNet. As is evident from these results, the proposed approach is an efficient automatic cardiac arrhythmia classification method and provides a reliable recognition system based on well-established CNN architectures instead of training a deep CNN from scratch.
{"title":"ECG CLASSIFICATION USING HIGHER ORDER SPECTRAL ESTIMATION AND DEEP LEARNING TECHNIQUES","authors":"Hiam Alquran, Ali Mohammad Alqudah, Isam Abu-Qasmieh, Alaa Al-Badarneh, S. Almashaqbeh","doi":"10.14311/nnw.2019.29.014","DOIUrl":"https://doi.org/10.14311/nnw.2019.29.014","url":null,"abstract":"Electrocardiogram (ECG) is one of the most important and effective tools in clinical routine to assess the cardiac arrhythmias. In this research higherorder spectral estimations, bispectrum and third-order cumulants, are evaluated, saved, and pre-trained using convolutional neural networks (CNN) algorithm. CNN is transferred in this study to carry out automatic ECG arrhythmia diagnostics after employing the higher-order spectral algorithms. Transfer learning strategies are applied on pre-trained convolutional neural network, namely AlexNet and GoogleNet, to carry out the final classification. Five different arrhythmias of ECG waveform are chosen from the MIT-BIH arrhythmia database to evaluate the proposed approach. The main contribution of this study is to utilize the pre-trained convolutional neural networks with a combination of higher-order spectral estimations of arrhythmias ECG signal to implement a reliable and applicable deep learning classification technique. The Highest average accuracy obtained is 97.8 % when using third cumulants and GoogleNet. As is evident from these results, the proposed approach is an efficient automatic cardiac arrhythmia classification method and provides a reliable recognition system based on well-established CNN architectures instead of training a deep CNN from scratch.","PeriodicalId":49765,"journal":{"name":"Neural Network World","volume":"1 1","pages":""},"PeriodicalIF":0.8,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67123276","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-01-01DOI: 10.14311/nnw.2019.29.026
J. Faber, M. Novák, M. Kovaljov, R. Chalupová, P. Bouchner
There is no methodical approach suitable for definition of the periodical or non-periodical, stationary or nonstationary curves of brain signals with a help of amplitude, frequency, phase etc. values. It is difficult to determinate the wave shape, i.e. the problem is how to solve the respective pattern recognition. Therefore, we tried to propose a simple method for praxis by help of measurement two main wave time components, interpreting a sinusoidal alpha wave as a triangle, where there is an anterior and a posterior part of wave ascending and descending abscissas in a hope that the sufficient measure are presented by the “legs” only or distances between upper and bottom peak of the wave. All the values of total ascendants are divided by all values of total descendants. For the method validity estimation it was made for this computation separately in two different psychical states – the relaxation and the calculation activity, both with eyes closed. Results are presented as quotient (quotus alpha) which means alpha waves symmetry. If the quotient is equal to 1, or is near to 1, is the alpha wave full or almost symmetrical. When the quotient is lower than 1 the ascendant is shorter than descendent, then alpha wave is asymmetric and has inclination to the left side. In contrary if the quotient is higher than 1 the ascendant is longer than descendent, alpha wave is again asymmetrical, but inclination is oriented to the right side. During mentation is usually quotient lower one and the ascendant is still more lover, alpha waves are sheer, the inclination to the left is more expressive.
{"title":"Dynamics of sinusoidal alpha waves asymmetry in brain electrical field","authors":"J. Faber, M. Novák, M. Kovaljov, R. Chalupová, P. Bouchner","doi":"10.14311/nnw.2019.29.026","DOIUrl":"https://doi.org/10.14311/nnw.2019.29.026","url":null,"abstract":"There is no methodical approach suitable for definition of the periodical or non-periodical, stationary or nonstationary curves of brain signals with a help of amplitude, frequency, phase etc. values. It is difficult to determinate the wave shape, i.e. the problem is how to solve the respective pattern recognition. Therefore, we tried to propose a simple method for praxis by help of measurement two main wave time components, interpreting a sinusoidal alpha wave as a triangle, where there is an anterior and a posterior part of wave ascending and descending abscissas in a hope that the sufficient measure are presented by the “legs” only or distances between upper and bottom peak of the wave. All the values of total ascendants are divided by all values of total descendants. For the method validity estimation it was made for this computation separately in two different psychical states – the relaxation and the calculation activity, both with eyes closed. Results are presented as quotient (quotus alpha) which means alpha waves symmetry. If the quotient is equal to 1, or is near to 1, is the alpha wave full or almost symmetrical. When the quotient is lower than 1 the ascendant is shorter than descendent, then alpha wave is asymmetric and has inclination to the left side. In contrary if the quotient is higher than 1 the ascendant is longer than descendent, alpha wave is again asymmetrical, but inclination is oriented to the right side. During mentation is usually quotient lower one and the ascendant is still more lover, alpha waves are sheer, the inclination to the left is more expressive.","PeriodicalId":49765,"journal":{"name":"Neural Network World","volume":"1 1","pages":""},"PeriodicalIF":0.8,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67123351","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-01-01DOI: 10.14311/nnw.2019.29.027
Qiang Li, Xiao-feng Liu
The dynamic stability assessment and prediction of a complex power system is a precondition to take the action of protecting control. This paper presents the four support vector machines (SVMs) with an improved genetic algorithm (GA) to compute their parameters automatically, that one SVM is used to simulate the tangent vector and the others for identifying the instable area. Besides, the GA was initialized by Meta-Learning method to enhance the performance and its optimal solution was selected by last test. Furthermore, a large network simplification was taken for reducing the amount of calculation and responding in real time. Study with the IEEE 118-bus test system indicated that the system status of a complex power system subjected a fault could be predicted based on this technique of the GA-SVM for simulating the tangent vector accurately. Besides, three binary SVM classifiers were trained to locate the instable area, and ranking the levels by the analysis of critical bus is help to management. Based on the test on the networks, the suggested approach can predict accurately with 98.87 % success rate and identify the fault area with 94.91 % success rate.
{"title":"Voltage stability assessment of complex power system based on GA-SVM","authors":"Qiang Li, Xiao-feng Liu","doi":"10.14311/nnw.2019.29.027","DOIUrl":"https://doi.org/10.14311/nnw.2019.29.027","url":null,"abstract":"The dynamic stability assessment and prediction of a complex power system is a precondition to take the action of protecting control. This paper presents the four support vector machines (SVMs) with an improved genetic algorithm (GA) to compute their parameters automatically, that one SVM is used to simulate the tangent vector and the others for identifying the instable area. Besides, the GA was initialized by Meta-Learning method to enhance the performance and its optimal solution was selected by last test. Furthermore, a large network simplification was taken for reducing the amount of calculation and responding in real time. Study with the IEEE 118-bus test system indicated that the system status of a complex power system subjected a fault could be predicted based on this technique of the GA-SVM for simulating the tangent vector accurately. Besides, three binary SVM classifiers were trained to locate the instable area, and ranking the levels by the analysis of critical bus is help to management. Based on the test on the networks, the suggested approach can predict accurately with 98.87 % success rate and identify the fault area with 94.91 % success rate.","PeriodicalId":49765,"journal":{"name":"Neural Network World","volume":"1 1","pages":""},"PeriodicalIF":0.8,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67123379","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","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：