Recently, a number of Extreme Learning Machine (ELM) based training algorithms have been introduced for training deep neural network structures. ELM based Auto-Encoder (ELM-AE) is one such algorithm that has been used for making multilayer structures and tuning parameters of each layer. In a simple ELM-AE training algorithm, the weights of the first layer are initialized randomly. This issue is a leading factor in producing reconstruction error. The frequent use of ELM-AE in deep network layers results in propagating such errors through deep structures and in decreasing performance as a consequent. In this paper, we introduce a multilayer structure and a new learning algorithm to train it that prevents error propagation. In order to boost the performance of the model, the parameters in the first layer are initialized by a novel type of ELM-AE called Repeated-AE (RAE) rather than by a random selection method. This RAE-based technique determines the parameters in the first layer far better than do the other ELM-AE existed methods. Next, a single hidden layer ELM is applied for handling the classification task. Experimental results for data classification show that the proposed method outperforms some other methods in terms of the average accuracy over all datasets by amounts of 4%, 26%, 17% and 31%. Eventually, so as to verify the performance of the proposed multilayer ELM-AE in application, we used this model to reconstruct images. The reconstructed images obtained by our approach appeared visually a lot better compared to those obtained by the other methods do.

Many columns that have been built in the past need to be repaired and strengthened for various reasons such as errors during construction, design errors, changes in the use of the building, changes in regulations, strong beam-weak column conditions and also injuries from accidents. Today, various methods for strengthening and improving structures, especially concrete columns, have become popular. One of the most common methods for reinforcing columns is to confine them using polymer fiber composites. The advantages of this method include increasing the axial and lateral bearing capacity of the column, increasing the compressive and tensile strength of the concrete member and increasing the ductility of the member. The mechanical properties of the concrete confined with FRP polymer fibers may be required for the purpose of design and rehabilitation. Artificial intelligence methods are among the modeling methods that have shown great power to coordinate with engineering problems. The aim of this study is to use multivariate adaptive regression spline (MARS) and extreme learning machine (ELM) artificial intelligence models to estimate the compressive strength of concrete in circular concrete columns confined with FRP polymer fibers. In addition, in order to improve the accuracy of these models, the particle swarm Optimization algorithm (PSO) is used in combination with these models and the accuracy of the models is evaluated to estimate the resistance. The results show that in overall, the used artificial intelligence models estimate the compressive strength of FRP- confined columns more accurately than the existing analytical models. In particular, the integrated MARS-PSO model has better performance compared to other models used, so that this model has a correlation coefficient of 0.9972 in the training stage and 0.9961 in the experimental stage. Also, the combination of PSO algorithm with MARS and ELM models improves their accuracy by 6.13 and 4.68 percent, respectively.

With a largeamount of multimedia content in the web, storage and retrieval of them by classical learning methods dealt with some major challenges like memory restriction. These limitations in some of learning algorithms like SVM and ANN is so serious that these algorithms cannot be employed in large-scale learning context. Kernel Extreme Learning Machine (KELM) algorithm is one of the powerful methods in machine learning. Learning phase of this method is based on constructing kernel matrix of labeled instances and calculating inverse of it. So, employing this method in large scale learning context with a lot of labeled instances is not feasible. In this research to overcome limitation of employing the KELM in large-scale multi-label learning, a new approach is proposed. The proposed approach is based on prototype selection in neighborhood of each training instance. By using the proposed approach, the size of training set is reduced. So, classical learning methods can be applied on reduced training set. Since multimedia contents are basically multi-label, the proposed prototype selection approach is based on multi-label domains like automatic image annotation. Experimental results on NUS-WIDE large-scale multi-label image set and three other versions include Object, Scene and Lite indicated the effectiveness of the proposed approach in solving the limitation of employing KELM method in large-scale multi-label learning.

Soil temperature (Ts) is a key factor controlling the soil physical, chemical and biological properties and processes and consequently affects agricultural crop productions. The objective of this study was to estimate Ts from meteorological data using different machine learning methods. For this purpose, meteorological data and soil temperature at different depths (5, 10, 20, 30, 50 and 100 cm from the soil surface) for 25 years (1994-2018) were collected from 17 synoptic stations in Khuzestan province, Iran. Air temperature, wind speed, relative humidity, evaporation, precipitation, and vapor pressure were used as inputs to train the models. Multiple Linear Regression (MLR), Multilayer Perceptron Neural Network (MLPNN) and Extreme Learning Machine (ELM) models were used to predict soil temperature from metrological data. The results indicated that all models predicted temperature of the top layer (0-30 cm) better than the ones in sublayers. on the other hand, by increasing soil depth the accuracy of the models diminished; so that, the best and worst Ts predictions were belong to 5 cm and 100 cm depth, respectively. The results revealed that MLR, MLPNN and ELM models provided desirable performance in modeling Ts at all depths, with R2 values ranging 0.700-0.864, 0.967-0.997, and 0.967-99, RMSE values ranging 2.557–2.873, 0.034–0.072, and 0.028–0.078 °C, and MAE values ranging 1.398–1.529, 0.023–0.063, and 0.023–0.065 °C, respectively. Overall, the results showed that MLPNN and ELM models had approximately similar performance and better accuracy than MLR model. However, because of the high computational speed of the ELM model, it is recommended to use MLPNN model for estimation of soil profile Ts.

Selection of training sites is an important and critical undertaking in the modeling procedure of mineral exploration targets using artificial intelligence approaches. This is because application of improper training algorithms results in exploration targeting models that carry bias and uncertainty. The present study aims to model exploration targets of porphyry copper mineralization in Chahar-Gonbad area, Kerman province, Iran, using artificial neural networks. In this regard, continuous weighted evidence maps of exploration criteria including proximity to intrusive contacts, fault density, multi-element geochemical signature and proximity to iron-oxide and argillic alterations were generated and applied as inputs to the neural network. Subsequently, 16 points with known mineral deposits and 16 points without mineralization were used to train the neural network through extreme learning algorithm. The ensuing exploration targeting model was compared with a model obtained by using geometric average integration method through prediction-area plot. The overall efficiency of the models are 0.34 and 0.27, respectively. Evaluation of the models demonstrated that the areas with high copper mineralization potential, marked as exploration targets, are in good conformity with known copper occurrences as well as with geological indicator features. Thus, the targets can be planned for further exploration programs.

In this research, optimized intelligent models were developed to design optimal sustainable concrete containing recycled Polyethylene Terephthalate (PET). For this aim, evolutionary Artificial Intelligence (AI) approach was implemented based on the integration of the Multivariate Adaptive Regression Splines (MARS) and Extreme Learning Machine (ELM) integrated with particle swarm optimization algorithm to investigate the strength behavior of sustainable concrete containing recycled polyethylene terephthalate-based fine aggregate. The experimental database consisting 273 records comprising mixture components at different ages are collected from published papers and optimal variables are identified using principal component analysis. The capability and efficiency of proposed model are validated through standalone MARS and ELM. Performance metrics indicated that proposed evolutionary formula-based models (MARS-PSO and ELM-PSO with the ((R= 0.902, RMSE=4.836 MPa and RSE=3.5) and (R= 0.900, RMSE=4.881 MPa and RSE=2.24), respectively) outperformed than other standalone AI models for CS prediction. Uncertainty analysis of the standalone and hybridized models is also applied using Monte-carlo simulation to prove that the hybridized multiscale model has less uncertainty in the prediction of the compressive strength compared to those benchmark models. The findings of the present paper presented the superiority of the model’s development in constructing reasonable and robustness evolutionary Model for formulation of CS of eco-friendly concrete containing recycled PET.

Selection and allocation in the resilient supply chain, when disruption threatens the supply chain, has become a strategic decision and the focus of many researches; increase in the applications of machine learning in supply chain studies has led to the emergence of faster and reliable decision-making methods, however, in few studies, machine learning has been used to deal with the problem of selecting and assigning suppliers to customers in resilient mode. The purpose of this research is to take a step towards solving this gap by using machine learning algorithms on real world data from the automotive supply chain in Iran. the performance data of 441 suppliers and 7 customers in 1401 was used. In this research, two clustering algorithms have been used to generate labels based on the concept of resilience capacity; Then, since the interpretability of the results was a priority, based on the labeling of the clusters by the experts, the decision tree was used to classify the suppliers based on their performance. The results showed the K-means tree performs better than the DBSCAN tree and criteria such as on-time delivery, capacity, production line stoppage, quality alert, logistics performance and quality performance are effective on suppliers' resilience.

In this paper, we propose an innovative classification method for tree bark classification and tree species identification. The proposed method consists of two steps. In the first step, we take the advantages of ILQP, a rotationally invariant, noise-resistant, and fully descriptive color texture feature extraction method. Then, in the second step, a new classification method called stacked mixture of ELM-based experts with a trainable gating network (stacked MEETG) is proposed. The proposed method is evaluated using the Trunk12, BarkTex, and AFF datasets. The performance of the proposed method on these three bark datasets shows that our approach provides better accuracy than other state-of-the-art methods.Our proposed method achieves an average classification accuracy of 92.79% (Trunk12), 92.54% (BarkTex), and 91.68% (AFF), respectively. Additionally, the results demonstrate that ILQP has better texture feature extraction capabilities than similar methods such as ILTP. Furthermore, stacked MEETG has shown a great influence on the classification accuracy.

Utilizing surrogate models based on artificial intelligence methods for detecting structural damages has attracted the attention of many researchers in recent decades. In this study, a new kernel based on Littlewood-Paley Wavelet (LPW) is proposed for Extreme Learning Machine (ELM) algorithm to improve the accuracy of detecting multiple damages in structural systems. ELM is used as metamodel (surrogate model) of exact finite element analysis of structures in order to efficiently reduce the computational cost through updating process. In the proposed two-step method, first a damage index, based on Frequency Response Function (FRF) of the structure, is used to identify the location of damages. In the second step, the severity of damages in identified elements is detected using ELM. In order to evaluate the efficacy of ELM, the results obtained from the proposed kernel were compared with other kernels proposed for ELM as well as Least Square Support Vector Machine algorithm. The solved numerical problems indicated that ELM algorithm accuracy in detecting structural damages is increased drastically in case of using LPW kernel.

Counting mitotic cellsis one of the main tasks involved in assessing breast cancer proliferation grade. Unfortunately, detection of mitoses present in the tissue is a challenging task. These cells have a wide variety of shape configurations and are sometimes very similar to apoptotic cells or external objects in the tissue sample. Utilizing image processing for automatic detection of mitotic cells is likely to reduce human errorandincreasegrading speed and performance.Most available mitosis detection methods extract many features from cells then classify cells using classic classifiers, or else, directly classify cells using neural networks. The former are fast but inaccurate methods, the latter being slow but accurate. In this work, we aim to present a simultaneously fast and accurate method based on a special type of neural networks, called ELM. After a pre-processing step, candidate cells are selected using thresholding and finding local maxima. An ELM is then directly trained with each cell image, without feature extraction. Our results indicate a considerable improvement over the status-quo. Our method also benefits from a very fast training time and test time.