Performance enhancement of extreme learning machine for multi-category sparse data classification problems

Sundaram, Suresh; Saraswathi, Saras; Natarajan, Sundararajan

doi:10.1016/j.engappai.2010.06.009

Cited by 133 publications

(65 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The ELM is a single hidden layer feed-forwarded ANN model, of which the input weights are calculated randomly, while the output weights are calculated analytically. Nondifferentiable or discrete activation functions can also be used in the hidden layer of the ELM, in addition to activation functions like sigmoidal, sine, Gaussian, and hard-limit [23]. Conventional feed-forwarded ANNs depend on some certain parameters like momentum or learning rate.…”

Section: Extreme Learning Machinementioning

confidence: 99%

“…Changing the momentum value may prevent the error from focusing on a local point, but it will not have any influence on the long learning process. The input weights and threshold values are produced randomly, but the output weights are obtained analytically in ELM [23]. The ELM network is the customized form of a single hidden layer feed-forwarded ANN model.…”

Section: Extreme Learning Machinementioning

confidence: 99%

See 1 more Smart Citation

A new intelligent classifier for breast cancer diagnosis based on a rough set and extreme learning machine: RS + ELM

Kaya¹

2013

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Breast cancer is one of the leading causes of death among women all around the world. Therefore, true and early diagnosis of breast cancer is an important problem. The rough set (RS) and extreme learning machine (ELM) methods were used collectively in this study for the diagnosis of breast cancer. The unnecessary attributes were discarded from the dataset by means of the RS approach. The classification process by means of ELM was performed using the remaining attributes. The Wisconsin Breast Cancer dataset (WBCD), derived from the University of California Irvine machine learning database, was used for the purpose of testing the proposed hybrid model and the success rate of the RS + ELM model was determined as 100%. Moreover, the most appropriate attributes for the diagnosis of breast cancer were determined from the WBCD in this study. It is considered that the proposed method will be useful in similar medical practices.

show abstract

Section: Extreme Learning Machinementioning

confidence: 99%

Section: Extreme Learning Machinementioning

confidence: 99%

A new intelligent classifier for breast cancer diagnosis based on a rough set and extreme learning machine: RS + ELM

Kaya¹

2013

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

show abstract

“…For input data set X = {x k }, let the desired outcome data from the network be Y = {y j } and the real outcome of the network be O = {o k } , where k ∈ [1, M ] represents the number of consequent input/output vectors. The mathematical description of the network having M neurons in the hidden layer, as given in Figure 4, can be expressed mathematically as [18]:…”

Section: Extreme Learning Machinementioning

confidence: 99%

“…The activation functions such as sigmoid, sine, Gaussian, and hard limit are used in the hidden layer; however, the linear activation function is used in the output layer. The nonderivative and discrete activation functions can be used in the ELM [18].…”

Section: Extreme Learning Machinementioning

confidence: 99%

EMG classification in obstructive sleep apnea syndrome and periodic limb movement syndrome patients by using wavelet packet transform and extreme learning machine

Sezgin¹

2015

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Electromyogram (EMG) signals, measured at the skin surface, provide crucial access to the muscle tones of a body. Some diseases, such as obstructive sleep apnea syndrome (OSAS) and periodic limb movement syndrome (PLMS), are closely associated with the electrical activity of muscle tones. In this paper, a hybrid model containing wavelet packet transform (WPT) plus an extreme learning machine (ELM) was proposed to classify EMG signals in OSAS and PLMS patients. At first, the WPT was used to extract the features of the EMG signal, and then these features were fed to the ELM classifier. The mean classification accuracy of the ELM was 96.85%. The obtained overall results were significant enough for specialists to diagnose OSAS and PLMS diseases. Furthermore, a remarkable relationship between OSAS and PLMS has been revealed.

show abstract

“…Yang et al [7] proposed evolutionary based ELM with differential evolution to balance the explorative power and exploitive power and to reduce the prediction time of original ELM. Suresh et al [8] presented a real-coded genetic algorithm to determine the input weights, optimal hidden neuron numbers, and bias values of ELM. In the study, proposed model gets a compact network with better generalization but it causes a higher computational effort.…”

Section: Introductionmentioning

confidence: 99%

Melez Aşırı Öğrenme Makinesi ve Türevi ile Hisse Senedi Fiyatı Tahmini

Göçken¹,

Özçalıcı²,

Boru³

et al. 2016

Çukurova Üniversitesi Mühendislik-Mimarlık Fakültesi Dergisi

View full text Add to dashboard Cite

Accurate and effective stock price prediction is appealing for investors due to the potential of obtaining a very high return. However, it is still a challenging task in the modern business world because of the complex, evolutionary, and nonlinear nature of stock market. Therefore, we proposed two hybrid models, which are Harmony Search (HS) based Extreme Learning Machine (ELM) that is denoted as HS-ELM and HS based Recurrent Extreme Learning Machine (RELM) that is represented as HS-RELM, to provide accurate and fast one-day ahead stock price prediction. This study provides a new direction in the field of stock price prediction and offers some suggestions on how to configure HS-ELM and HS-RELM for performing stock price prediction, with an application on stocks listed in BIST50 Index. The results of the performance measures show that although both proposed models are very helpful for the practical applicability of the stock market, HS-RELM model is more powerful than HS-ELM model. Keywords

show abstract

Performance enhancement of extreme learning machine for multi-category sparse data classification problems

Cited by 133 publications

References 8 publications

A new intelligent classifier for breast cancer diagnosis based on a rough set and extreme learning machine: RS + ELM

A new intelligent classifier for breast cancer diagnosis based on a rough set and extreme learning machine: RS + ELM

EMG classification in obstructive sleep apnea syndrome and periodic limb movement syndrome patients by using wavelet packet transform and extreme learning machine

Melez Aşırı Öğrenme Makinesi ve Türevi ile Hisse Senedi Fiyatı Tahmini

Contact Info

Product

Resources

About