DOCSLIB.ORG

Study Comparison Backpropogation, Support Vector Machine, and Extreme Learning Machine for Bioinformatics Data

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Study Comparison Backpropogation, Support Vector Machine, and Extreme Learning Machine for Bioinformatics Data Jurnal Ilmu Komputer dan Informasi (Journal of Computer Science and Information). 8/1 (2015), 53-59 DOI: http://dx.doi.org/10.21609/jiki.v8i1.284 STUDY COMPARISON BACKPROPOGATION, SUPPORT VECTOR MACHINE, AND EXTREME LEARNING MACHINE FOR BIOINFORMATICS DATA Umi Mahdiyah1, M. Isa Irawan1, and Elly Matul Imah2 1Faculty of Mathematics and Science, Institut Teknologi Sepuluh Nopember, Jl. Arief Rahman Hakim, Surabaya, 60111, Indonesia 2 Mathematics Department, Universitas Negeri Surabaya, Jl. Ketintang, Surabaya, 60231, Indonesia E-mail: [email protected] Abstract A successful understanding on how to make computers learn would open up many new uses of computers and new levels of competence and customization. A detailed understanding on inform- ation- processing algorithms for machine learning might lead to a better understanding of human learning abilities and disabilities. There are many type of machine learning that we know, which includes Backpropagation (BP), Extreme Learning Machine (ELM), and Support Vector Machine (SVM). This research uses five data that have several characteristics. The result of this research is all the three investigated models offer comparable classification accuracies. This research has three type conclusions, the best performance in accuracy is BP, the best performance in stability is SVM and the best performance in CPU time is ELM for bioinformatics data. Keywords: Machine Learning, Backpropagation, Extreme Learning Machine, Support Vector Machine, Bioinformatics Abstrak Keberhasilan pemahaman tentang bagaimana membuat komputer belajar akan membuka banyak manfaat baru dari komputer. Sebuah pemahaman yang rinci tentang algoritma pengolahan informasi untuk pembelajaran mesin dapat membuat pemahaman yang sebaik kemampuan belajar manusia. Banyak jenis pembelajaran mesin yang kita tahu, beberapa diantaranya adalah Backpropagation (BP), Extreme Learning Machine (ELM), dan Support Vector Machine (SVM). Penelitian ini menggunakan lima data yang memiliki beberapa karakteristik. Hasil penelitian ini, dari ketiga model yang diamati memberikan akurasi klasifikasi yang sebanding. Penelitian ini memiliki tiga kesimpulan, yang terbaik dalam akurasi adalah BP, yang terbaik dalam stabilitas adalah SVM dan CPU time terbaik adalah ELM untuk data bioinformatika. Kata Kunci: Machine Learning, Backpropagation, Extreme Learning Machine, Support Vector Machine, Bioinformatika 1. Introduction gradient descent type algorithm, which has con- nection parameter for each step or iteration. But, A successful understanding of how to make com- this algorithm can provide harmony result bet- puters learn would open up many new uses of ween "network capability" to recognize the pat- computers and new levels of competence and cus- terns which used for training and "network capa- tomization. And a detailed understanding of infor- bility" to respond correctly to the input patterns mation-processing algorithms for machine learn- that similar (but not equal) to the training pattern. ing might lead to a better understanding of human Then, Backpropagation algorithm has limit classi- learning abilities (and disabilities) as well [1]. fication accuracy, because if the output value is Many type of machine learning that we know, so- different from the target value, an error will be me of them are Backpropagation (BP), Extreme calculated, and then taken from the output layer to Learning Machine (ELM), and Support Vector the input layer (Backpropagation process). Machine (SVM). Until now, many researchers have developed First Machine Learning is backpropagation. and implemented the BP algorithm [3-7]. Imple- Backpropagation was initially formulated by We- mentation BP for classification was implemented bros in 1974, which was later modified by Rumel- in most problem as bioinformatics, biomedical, hart and McClelland [2]. Backpropagation is the chemistry, art, environment, etc. Several research 53 54 Jurnal Ilmu Komputer dan Informasi (Journal of Computer Science and Information), Volume 8, Issue 1, February 2015 use this algorithm because BP has minimum error 2. Methods and effective for some problem, especially in this study for classification problem. Backpropagation The next powerful machine learning is Sup- port Vector Machines (SVM). Support vector ma- Backpropagation is one of many supervised ma- chines (SVM) are a group of supervised learning chine learning. Backpropagation is the gradient methods that can be applied to classification or re- descent type algorithm .This algorithm has two gression. This Machine learning is different from phases for processing data. First phase, input vec- Machine Learning in ANN. In SVM when train- tor given to input layer, continued to hidden layer ing process, SVM isn’t training all data, but this then finding output value in output layer. Second algorithm just training support vector data. Based phase, if the output value is different from the tar- idea from this algorithm is optimization margin get value, an error will be calculated, and then ta- hyper plane. Although SVM need big memory ken from the output layer to the input layer (Back- and need long time to process data, but many re- propagation process) [19]. searcher use SVM to solve some problem [8-11], In BP, transfer function must fulfill several because this machine learning have high perfor- conditions: continue differentiable, not descend- mance. ing function. In this paper we use sigmoid binary Extreme Learning Machine is one of a new function as defined in equation(1) and its deriva- learning algorithm in neural networks, which has tion as in equation(2). the Single-hidden Layer feed-forward Network (SLFN). ELM has a very fast learning capability 1 ( ) = (1) and training small error [12]. First ELM was in- 1 + troduced by Huang in 2004 as Single-hidden Lay- − er Feed-forward Network. ELM was made to ov- ( ) = ( ) 1 ( ) (2) ercome the weaknesses of the feed-forward neural ′ networks problem that learning speed. Tradition- Backpropagati on training � − follow � these step ally, feed-forward neural network using gradient- [20]: a) Initialize neuron’s weight with random based learning algorithm for training, as well as number. all the parameters (input weight and hidden bias) Forward propagation are determined by iterative network, to solve that b) Each input layer neuron receives the inputs and problem, Extreme Learning Machine using mini- passes it to the connected hidden layer’s neuron. mum norm least-squares (LS) solution of SLFNs. c) Compute output value from hidden layer , = Unlike the traditional function approximation 1,2 … , , with the weight is that connected theories which require to adjusted input weights input layer and hidden layer using equation(3) and and hidden layer biases, input weights and hidden equation(4) . layer biases can be randomly assigned if only the activation function is infinitely differentiable [13]. So, ELM can be faster than prior the neural net- = + (3) work algorithm previously. ELM has been applied and developed in various fields [14-18]. Up until 0 � now this algorithm is developed, the main reason =1 1 many research use this algorithm because this al- = = (4) 1 + gorithm is simple and faster than several algori- − thm in ANN. � � If the hidden layer end, we continue to output lay- In this paper, we will compare three algori- er. d) Compute all output values from output layer thm for classification several bioinformatics data as equation(5) and equation(6). that have some criteria. Dataset was taken from UCI Machine Learning. We use general BP, SVM using linear kernel and general ELM. We choose = + (5) Backpropagation (BP), Extreme Learning Machi- ne (ELM), and Support Vector Machine (SVM) 0 � because that all machine learning have good per- =1 1 = ( _ ) = (6) formance for classification [2-18]. The rest of the 1 + _ paper is organized as follows. In Section 2, we − present the detail of BP, SVM, and ELM. Follow- Backpropagation ing that, Section 3 provides experimental results e) Compute the output layer’s error factor based and discussions. Finally, we draw conclusions in on the error in each of the output layer’s neuron Section 4. using equation(7) and equation(8). Umi Mahdiyah, et al., Study Comparison Backpropagation 55 TABLE 1 Forward propagation KERNEL FUNCTION Kernel Function 1 1 y 1 Linear ( , ) = . Polynomial ( , ) = ( . + ) 2 x1 1 y2 2 …. …. Gaussian RBF ( , ) = exp 2. 2 x i 3 −‖ −‖ 2 k yk Sigmoid(tangent (, ) = �( ( . )2+ �) …. hyperbolic) ℎ 1 xn n j l yn Multiquadratic ( , ) = Invers | | + w ij 2 wjk 2 m �� − � Back propagation m Figure 1. Backpropagation architecture. _ = ( ) (7) Positive margin w.xi+b=1 ′ =− _ � � (8) Negative margin Hyperplane f) Compute the ∆hidden layer’s∝ error factor based w.xi+b=-1 w.xi+b=0 on error in each of the hidden layer’s neuron using equation(9) to equation(11). Negative class Positive class Figure 2. Margin hyperplane SVM. _ = (9) classes can be obtained, that called linear hyper � plane classifiers [21]. =1 _ = _ ( _ ) (10) This is step of support vector machine algo- rithm: Given training data set D and q feature of = ′_ (11) each data, there are , such that formulated in eq- uation(14) and equation(15). g) The last step count∆ all∝ change of weight of neu- Φ rons using equation(12) and equation(13). : (14) ( ) = ( ) + (12) Φ →( ) (15) ( ) = ( ) + ∆ (13) Where r is new feature set →Φ that result from map- ping D. while x is training data, which , , …, is
Load more

Recommended publications
  • Getting Started with Machine Learning
    Getting Started with Machine Learning CSC131 The Beauty & Joy of Computing Cornell College 600 First Street SW Mount Vernon, Iowa 52314 September 2018 ii Contents 1 Applications: where machine learning is helping1 1.1 Sheldon Branch............................1 1.2 Bram Dedrick.............................4 1.3 Tony Ferenzi.............................5 1.3.1 Benefits of Machine Learning................5 1.4 William Golden............................7 1.4.1 Humans: The Teachers of Technology...........7 1.5 Yuan Hong..............................9 1.6 Easton Jensen............................. 11 1.7 Rodrigo Martinez........................... 13 1.7.1 Machine Learning in Medicine............... 13 1.8 Matt Morrical............................. 15 1.9 Ella Nelson.............................. 16 1.10 Koichi Okazaki............................ 17 1.11 Jakob Orel.............................. 19 1.12 Marcellus Parks............................ 20 1.13 Lydia Sanchez............................. 22 1.14 Tiff Serra-Pichardo.......................... 24 1.15 Austin Stala.............................. 25 1.16 Nicole Trenholm........................... 26 1.17 Maddy Weaver............................ 28 1.18 Peter Weber.............................. 29 iii iv CONTENTS 2 Recommendations: How to learn more about machine learning 31 2.1 Sheldon Branch............................ 31 2.1.1 Course 1: Machine Learning................. 31 2.1.2 Course 2: Robotics: Vision Intelligence and Machine Learn- ing............................... 33 2.1.3 Course
    [Show full text]
  • Deep Hashing Using an Extreme Learning Machine with Convolutional Networks
    i \1-Zeng" | 2018/2/2 | 23:57 | page 133 | #1 i i i Communications in Information and Systems Volume 17, Number 3, 133{146, 2017 Deep hashing using an extreme learning machine with convolutional networks Zhiyong Zeng∗, Shiqi Dai, Yunsong Li, Dunyu Chen In this paper, we present a deep hashing approach for large scale image search. It is different from most existing deep hash learn- ing methods which use convolutional neural networks (CNN) to execute feature extraction to learn binary codes. These methods could achieve excellent results, but they depend on an extreme huge and complex networks. We combine an extreme learning ma- chine (ELM) with convolutional neural networks to speed up the training of deep learning methods. In contrast to existing deep hashing approaches, our method leads to faster and more accurate feature learning. Meanwhile, it improves the generalization ability of deep hashing. Experiments on large scale image datasets demon- strate that the proposed approach can achieve better results with state-of-the-art methods with much less complexity. 1. Introduction With the growing of multimedia data, fast and effective search technique has become a hot research topic. Among existing search techniques, hashing is one of the most important retrieval techniques due to its fast query speed and low memory cost. Hashing methods can be divided into two categories: data-independent [1{3], data-dependent [4{8]. For the first category, hash functions random generated are first used to map data into feature space and then binarization is carried out. Representative methods of this category are locality sensitive hashing (LSH) [1] and its variants [2, 3].
    [Show full text]
  • Convolutional Neural Network Based on Extreme Learning Machine for Maritime Ships Recognition in Infrared Images
    sensors Article Convolutional Neural Network Based on Extreme Learning Machine for Maritime Ships Recognition in Infrared Images Atmane Khellal 1,*, Hongbin Ma 1,2,* and Qing Fei 1,2 1 School of Automation, Beijing Institute of Technology, Beijing 100081, China; [email protected] 2 State Key Laboratory of Intelligent Control and Decision of Complex Systems, Beijing Institute of Technology, Beijing 100081, China * Correspondence: [email protected] (A.K.); [email protected] (H.M.); Tel.: +86-132-6107-8800 (A.K.); +86-152-1065-8048 (H.M.) Received: 1 April 2018; Accepted: 7 May 2018; Published: 9 May 2018 Abstract: The success of Deep Learning models, notably convolutional neural networks (CNNs), makes them the favorable solution for object recognition systems in both visible and infrared domains. However, the lack of training data in the case of maritime ships research leads to poor performance due to the problem of overfitting. In addition, the back-propagation algorithm used to train CNN is very slow and requires tuning many hyperparameters. To overcome these weaknesses, we introduce a new approach fully based on Extreme Learning Machine (ELM) to learn useful CNN features and perform a fast and accurate classification, which is suitable for infrared-based recognition systems. The proposed approach combines an ELM based learning algorithm to train CNN for discriminative features extraction and an ELM based ensemble for classification. The experimental results on VAIS dataset, which is the largest dataset of maritime ships, confirm that the proposed approach outperforms the state-of-the-art models in term of generalization performance and training speed.
    [Show full text]
  • A Real-Time Sequential Deep Extreme Learning Machine Cybersecurity Intrusion Detection System
    Computers, Materials & Continua Tech Science Press DOI:10.32604/cmc.2020.013910 Article A Real-Time Sequential Deep Extreme Learning Machine Cybersecurity Intrusion Detection System Amir Haider1, Muhammad Adnan Khan2, Abdur Rehman3, Muhib Ur Rahman4 and Hyung Seok Kim1,* 1Department of Intelligent Mechatronics Engineering, Sejong University, Seoul, 05006, Korea 2Department of Computer Science, Lahore Garrison University, Lahore, 54000, Pakistan 3School of Computer Science, National College of Business Administration & Economics, Lahore, 54000, Pakistan 4Department of Electrical Engineering, Polytechnique Montreal, Montreal, QC H3T 1J4, Canada ÃCorresponding Author: Hyung Seok Kim. Email: [email protected] Received: 26 August 2020; Accepted: 28 September 2020 Abstract: In recent years, cybersecurity has attracted significant interest due to the rapid growth of the Internet of Things (IoT) and the widespread development of computer infrastructure and systems. It is thus becoming particularly necessary to identify cyber-attacks or irregularities in the system and develop an efficient intru- sion detection framework that is integral to security. Researchers have worked on developing intrusion detection models that depend on machine learning (ML) methods to address these security problems. An intelligent intrusion detection device powered by data can exploit artificial intelligence (AI), and especially ML, techniques. Accordingly, we propose in this article an intrusion detection model based on a Real-Time Sequential Deep Extreme Learning Machine Cyber- security Intrusion Detection System (RTS-DELM-CSIDS) security model. The proposed model initially determines the rating of security aspects contributing to their significance and then develops a comprehensive intrusion detection frame- work focused on the essential characteristics. Furthermore, we investigated the feasibility of our proposed RTS-DELM-CSIDS framework by performing dataset evaluations and calculating accuracy parameters to validate.
    [Show full text]
  • Road Traffic Prediction Model Using Extreme Learning Machine: the Case Study of Tangier, Morocco
    information Article Road Traffic Prediction Model Using Extreme Learning Machine: The Case Study of Tangier, Morocco Mouna Jiber *, Abdelilah Mbarek *, Ali Yahyaouy , My Abdelouahed Sabri and Jaouad Boumhidi Department of Computer Science, Faculty of Sciences Dhar El Mahraz, Sidi Mohamed Ben Abdellah University, Fez 30003, Morocco; [email protected] (A.Y.); [email protected] (M.A.S.); [email protected] (J.B.) * Correspondence: [email protected] (M.J.); [email protected] (A.M.) Received: 29 September 2020; Accepted: 10 November 2020; Published: 24 November 2020 Abstract: An efficient and credible approach to road traffic management and prediction is a crucial aspect in the Intelligent Transportation Systems (ITS). It can strongly influence the development of road structures and projects. It is also essential for route planning and traffic regulations. In this paper, we propose a hybrid model that combines extreme learning machine (ELM) and ensemble-based techniques to predict the future hourly traffic of a road section in Tangier, a city in the north of Morocco. The model was applied to a real-world historical data set extracted from fixed sensors over a 5-years period. Our approach is based on a type of Single hidden Layer Feed-forward Neural Network (SLFN) known for being a high-speed machine learning algorithm. The model was, then, compared to other well-known algorithms in the prediction literature. Experimental results demonstrated that, according to the most commonly used criteria of error measurements (RMSE, MAE, and MAPE), our model is performing better in terms of prediction accuracy.
    [Show full text]
  • Research Article Deep Extreme Learning Machine and Its Application in EEG Classification
    Hindawi Publishing Corporation Mathematical Problems in Engineering Volume 2015, Article ID 129021, 11 pages http://dx.doi.org/10.1155/2015/129021 Research Article Deep Extreme Learning Machine and Its Application in EEG Classification Shifei Ding,1,2 Nan Zhang,1,2 Xinzheng Xu,1,2 Lili Guo,1,2 and Jian Zhang1,2 1 School of Computer Science and Technology, China University of Mining and Technology, Xuzhou 221116, China 2Key Laboratory of Intelligent Information Processing, Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China Correspondence should be addressed to Shifei Ding; [email protected] Received 26 August 2014; Revised 4 November 2014; Accepted 12 November 2014 Academic Editor: Amaury Lendasse Copyright © 2015 Shifei Ding et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Recently, deep learning has aroused wide interest in machine learning fields. Deep learning is a multilayer perceptron artificial neural network algorithm. Deep learning has the advantage of approximating the complicated function and alleviating the optimization difficulty associated with deep models. Multilayer extreme learning machine (MLELM) is a learning algorithm of an artificial neural network which takes advantages of deep learning and extreme learning machine. Not only does MLELM approximate the complicated function but it also does not need to iterate during the training process. We combining with MLELM and extreme learning machine with kernel (KELM) put forward deep extreme learning machine (DELM) and apply it to EEG classification in this paper.
    [Show full text]
  • Chaotic Ensemble of Online Recurrent Extreme Learning Machine for Temperature Prediction of Control Moment Gyroscopes
    sensors Letter Chaotic Ensemble of Online Recurrent Extreme Learning Machine for Temperature Prediction of Control Moment Gyroscopes Luhang Liu 1,2 , Qiang Zhang 2, Dazhong Wei 2, Gang Li 2, Hao Wu 2, Zhipeng Wang 3 , Baozhu Guo 1,* and Jiyang Zhang 2 1 China Aerospace Academy of Systems Science and Engineering, Beijing 100037, China; [email protected] 2 Beijing Institute of Control Engineering, Beijing 100094, China; [email protected] (Q.Z.); [email protected] (D.W.); [email protected] (G.L.); [email protected] (H.W.); [email protected] (J.Z.) 3 State Key Lab of Rail Traffic Control & Safety, Beijing Jiaotong University, Beijing 100044, China; [email protected] * Correspondence: [email protected] Received: 30 July 2020; Accepted: 19 August 2020; Published: 25 August 2020 Abstract: Control moment gyroscopes (CMG) are crucial components in spacecrafts. Since the anomaly of bearing temperature of the CMG shows apparent correlation with nearly all critical fault modes, temperature prediction is of great importance for health management of CMGs. However, due to the complicity of thermal environment on orbit, the temperature signal of the CMG has strong intrinsic nonlinearity and chaotic characteristics. Therefore, it is crucial to study temperature prediction under the framework of chaos time series theory. There are also several other challenges including poor data quality, large individual differences and difficulty in processing streaming data. To overcome these issues, we propose a new method named Chaotic Ensemble of Online Recurrent Extreme Learning Machine (CE-ORELM) for temperature prediction of control moment gyroscopes. By means of the CE-ORELM model, this proposed method is capable of dynamic prediction of temperature.
    [Show full text]
  • Optimized Extreme Learning Machine for Forecasting Confirmed Cases of COVID-19
    Received: November 16, 2020. Revised: February 3, 2021. 484 Optimized Extreme Learning Machine for Forecasting Confirmed Cases of COVID-19 Ahmed Mudheher Hasan1* Aseel Ghazi Mahmoud2 Zainab Mudheher Hasan3 1Control and Systems Engineering Department, University of Technology-Iraq, Baghdad, Iraq 2College of Nursing, University of Baghdad, Iraq 3Ministry of Health and Environment of Iraq, Baghdad health directorate/Rusafa, Iraq * Corresponding author’s Email: [email protected] Abstract: Recently, a new challenge to the researchers has been emerged due to the spread of a new outbreak called novel corona-virus (COVID-19) to portend the confirmed cases. Since discovering the first certain cases in Wuhan, China, the COVID-19 has been widely spreading and expanding to other provinces and other countries via travellers into various countries around the world. COVID-19 virus is not a problem of only developing countries, but also of developed countries. Artificial Intelligence (AI) techniques can be profitable to predict such; parameters, risks and influence of an outbreak. Therefore, accurate prognosis can be helpful to control the spread of such viruses and provide crucial information for identifying the type of virus interventions and intensity. Here are develop an intelligent model depicting COVID-19 transmission and resulting confirmed cases. The epidemic curve of COVID-19 cases was modelled. The main key idea of predicting the confirmed cases is based on two factors 1) cumulative number of confirmed cases and 2) the daily confirmed cases instead of using only one factor as previous research. In this study, a comparison between different intelligent techniques has been conducted. To assess the effectiveness of these intelligent models, a recorded data of 5 months has been used for training in various countries.
    [Show full text]
  • Constrained Extreme Learning Machines: a Study on Classification Cases
    > REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 1 Constrained Extreme Learning Machines: A Study on Classification Cases Wentao Zhu, Jun Miao, and Laiyun Qing (LM) method [5], dynamically network construction [6], Abstract—Extreme learning machine (ELM) is an extremely evolutionary algorithms [7] and generic optimization [8], fast learning method and has a powerful performance for pattern the enhanced methods require heavy computation or recognition tasks proven by enormous researches and engineers. cannot obtain a global optimal solution. However, its good generalization ability is built on large numbers 2. Optimization based learning methods. One of the most of hidden neurons, which is not beneficial to real time response in the test process. In this paper, we proposed new ways, named popular optimization based SLFNs is Support Vector “constrained extreme learning machines” (CELMs), to randomly Machine (SVM) [9]. The objective function of SVM is select hidden neurons based on sample distribution. Compared to to optimize the weights for maximum margin completely random selection of hidden nodes in ELM, the CELMs corresponding to structural risk minimization. The randomly select hidden nodes from the constrained vector space solution of SVM can be obtained by convex containing some basic combinations of original sample vectors. optimization methods in the dual problem space and is The experimental results show that the CELMs have better the global optimal solution. SVM is a very popular generalization ability than traditional ELM, SVM and some other method attracting many researchers [10]. related methods. Additionally, the CELMs have a similar fast learning speed as ELM. 3.
    [Show full text]
  • 92 a Survey on Deep Learning: Algorithms, Techniques, And
    A Survey on Deep Learning: Algorithms, Techniques, and Applications SAMIRA POUYANFAR, Florida International University SAAD SADIQ and YILIN YAN, University of Miami HAIMAN TIAN, Florida International University YUDONG TAO, University of Miami MARIA PRESA REYES, Florida International University MEI-LING SHYU, University of Miami SHU-CHING CHEN and S. S. IYENGAR, Florida International University The field of machine learning is witnessing its golden era as deep learning slowly becomes the leaderin this domain. Deep learning uses multiple layers to represent the abstractions of data to build computational models. Some key enabler deep learning algorithms such as generative adversarial networks, convolutional neural networks, and model transfers have completely changed our perception of information processing. However, there exists an aperture of understanding behind this tremendously fast-paced domain, because it was never previously represented from a multiscope perspective. The lack of core understanding renders these powerful methods as black-box machines that inhibit development at a fundamental level. Moreover, deep learning has repeatedly been perceived as a silver bullet to all stumbling blocks in machine learning, which is far from the truth. This article presents a comprehensive review of historical and recent state-of-the- art approaches in visual, audio, and text processing; social network analysis; and natural language processing, followed by the in-depth analysis on pivoting and groundbreaking advances in deep learning applications.
    [Show full text]
  • Redalyc.Extreme Learning Machine to Analyze the Level of Default In
    Revista de Métodos Cuantitativos para la Economía y la Empresa E-ISSN: 1886-516X [email protected] Universidad Pablo de Olavide España Montero-Romero, Teresa; López-Martín, María del Carmen; Becerra-Alonso, David; Martínez- Estudillo, Francisco José Extreme Learning Machine to Analyze the Level of Default in Spanish Deposit Institutions Revista de Métodos Cuantitativos para la Economía y la Empresa, vol. 13, 2012, pp. 3-23 Universidad Pablo de Olavide Sevilla, España Available in: http://www.redalyc.org/articulo.oa?id=233124421001 How to cite Complete issue Scientific Information System More information about this article Network of Scientific Journals from Latin America, the Caribbean, Spain and Portugal Journal's homepage in redalyc.org Non-profit academic project, developed under the open access initiative REVISTA DE METODOS¶ CUANTITATIVOS PARA LA ECONOM¶IA Y LA EMPRESA (13). P¶aginas3{23. Junio de 2012. ISSN: 1886-516X. D.L: SE-2927-06. URL: http://www.upo.es/RevMetCuant/art.php?id=56 Extreme Learning Machine to Analyze the Level of Default in Spanish Deposit Institutions Montero-Romero, Teresa Department of Management and Quantitative Methods, ETEA, C¶ordoba(Spain) Correo electr¶onico: [email protected] Lopez-Mart¶ ¶³n, Mar¶³a del Carmen Department of Economics, Legal Sciences and Sociology, ETEA, C¶ordoba(Spain) Correo electr¶onico: [email protected] Becerra-Alonso, David Department of Management and Quantitative Methods, ETEA, C¶ordoba(Spain) Correo electr¶onico: [email protected] Mart¶³nez-Estudillo, Francisco Jose¶ Department of Management and Quantitative Methods, ETEA, C¶ordoba(Spain) Correo electr¶onico: [email protected] ABSTRACT The level of default in ¯nancial institutions is a key piece of information in the activity of these organizations and reveals their level of risk.
    [Show full text]
  • Extreme Learning Machine: Theory and Applications
    ARTICLE IN PRESS Neurocomputing 70 (2006) 489–501 www.elsevier.com/locate/neucom Extreme learning machine: Theory and applications Guang-Bin HuangÃ, Qin-Yu Zhu, Chee-Kheong Siew School of Electrical and Electronic Engineering, NanyangTechnological University, Nanyang Avenue, Singapore 639798, Singapore Received 27 January 2005; received in revised form 1 December 2005; accepted 3 December 2005 Communicated by J. Tin-Yau Kwok Available online 16 May 2006 Abstract It is clear that the learning speed of feedforward neural networks is in general far slower than required and it has been a major bottleneck in their applications for past decades. Two key reasons behind may be: (1) the slow gradient-based learning algorithms are extensively used to train neural networks, and (2) all the parameters of the networks are tuned iteratively by using such learning algorithms. Unlike these conventional implementations, this paper proposes a new learning algorithm called extreme learning machine (ELM) for single-hidden layer feedforward neural networks (SLFNs) which randomly chooses hidden nodes and analytically determines the output weights of SLFNs. In theory, this algorithm tends to provide good generalization performance at extremely fast learning speed. The experimental results based on a few artificial and real benchmark function approximation and classification problems including very large complex applications show that the new algorithm can produce good generalization performance in most cases and can learn thousands of times faster than conventional popular learning algorithms for feedforward neural networks.1 r 2006 Elsevier B.V. All rights reserved. Keywords: Feedforward neural networks; Back-propagation algorithm; Extreme learning machine; Support vector machine; Real-time learning; Random node 1.
    [Show full text]