The increase of the worldwide installed photovoltaic (PV) capacity and the intermittent nature of the solar resource highlights the importance of power forecasting for the grid integration of the ...technology. This study compares 24 machine learning models for deterministic day-ahead power forecasting based on numerical weather predictions (NWP), tested for two-year-long 15-min resolution datasets of 16 PV plants in Hungary. The effects of the predictor selection and the benefits of the hyperparameter tuning are also evaluated. The results show that the two most accurate models are kernel ridge regression and multilayer perceptron with an up to 44.6% forecast skill score over persistence. Supplementing the basic NWP data with Sun position angles and statistically processed irradiance values as the inputs of the learning models results in a 13.1% decrease of the root mean square error (RMSE), which underlines the importance of the predictor selection. The hyperparameter tuning is essential to exploit the full potential of the models, especially for the less robust models, which are prone to under or overfitting without proper tuning. The overall best forecasts have a 13.9% lower RMSE compared to the baseline scenario of using linear regression. Moreover, the power forecasts based on only daily average irradiance forecasts and the Sun position angles have only a 1.5% higher RMSE than the best scenario, which demonstrates the effectiveness of machine learning even for limited data availability. The results of this paper can support both researchers and practitioners in constructing the best data-driven techniques for NWP-based PV power forecasting.
•24 machine learning models tested for day-ahead photovoltaic power forecasting.•Kernel ridge and multilayer perceptron are the overall most accurate models.•Predictor selection is even more important than model selection.•Hyperparameter optimization is essential for the highest accuracy.•Up to 13.9% RMSE improvement over a baseline linear regression model.
Mobile cloud computing based stroke healthcare system Karaca, Yeliz; Moonis, Majaz; Zhang, Yu-Dong ...
International journal of information management,
April 2019, 2019-04-00, 20190401, Letnik:
45
Journal Article
Recenzirano
MLP in Stroke Diagnosis Healthcare System seems to be crucial concerning the use of mobile cloud computing for the management of stroke data in healthcare systems. In general, life quality of the ...patients will be improved since they will have more agency over their disease. Once they enter the analysis results on the system set up, they can have direct access to the relevant data so they can have agency over their course of disease, which render them more aware and informed patients. The secondary benefit would be concerned with the psychological aspect since agency and more awareness would help reduce anxiety and concerns regarding the disease.
•The study suggests a relevant solution that is Knowledge-based Oriented Stroke Diagnosis Healthcare Application for cardioembolic/cryptogenic stroke patients.•The study includes big data application, supported by ANN, it has been used on all mobile phone running Android OS with cloud computing for the first time concerning Stroke dataset.•The proposed system will enhance accessibility to accurate health data and analysis for the patients,•The system will also improve the accuracy rate since classification is performed by ANN clustering affordance.•The system will serve to provide economic benefits since the information is accessed directly through mobile system of the user.•Overall, the life quality of the Stroke patients will increase as a result of using the proposed novel system.
Information technology has recently seen a huge progress in innovative healthcare technologies that rendered healthcare data bigger. Connectivity on 7/24 basis between human to device and device to device have a crucial role in individuals’ lives. Therefore, Mobile Cloud System (MCC) has become an indispensable tool. Parallel with the rapid developments in the Internet of Things, convergence has become an important issue. Our proposed method, accordingly, can be converged with mobile-cloud environments with cloud computing in handling healthcare information. This study uses Virtual Dedicated Server (VDS) as 4 VCPU and 8 GB RAM and proposes a model based on the Android based mobile phones for stroke patients with cardioembolic (689) and cryptogenic (528) subtypes. The system set up through this study has two basic application elements which are mobile application and server application. Artificial Neural Network (ANN) module is beneficial for classifying the two stroke subtypes while server application is used for saving the data from the patients. Accordingly, our model guarantees availability, security, and scalability as a system for stroke patients applying Stroke dataset for ANN algorithm, Multilayer Perceptron Algorithm (MLP), which has been done for the first time in literature with big data in this scope. The main contributions are: (1) The outcomes will display an individual unique social insurance framework. (2) The outcomes will be utilized for the distinguishing proof of stroke-related data to be gathered by cell phones that are Android based. (3) Stroke patients will find out about their condition of well-being through an ANN application programming interface, which will provide a sort of organization for the patients. Overall, an efficient and user-friendly stroke determination human services framework has been presented through this Healthcare System for patients.
In the data‐driven society, fidelity and accuracy of automatic decisions behind the scene rely fundamentally on a solid data or imaging acquisition system. However, conventional microwave imagers are ...inadequate relating to their resolution and noise capability, mainly due to the limited aperture size and rigid working principle. Here, a programmable metasurface imager with high‐resolution and anti‐interference performance is proposed. By implementing the structure of multilayer perceptron network in the analog domain, the metasurface‐based microwave imager intelligently adapts to different datasets through illuminating a set of designed scattering patterns that mimic the feature patterns. A prototype imager system working at microwave frequency is designed and fabricated. The accuracy rate rises by 18% under the classification task of MNIST dataset, with a decline in the reconstruction imaging error. The authors experimentally demonstrate that the resolution to distinguish strip patterns goes beyond to one‐fifth of the equivalent wavelength on the target plane.
The article reports a novel imaging system architecture based on metasurface and machine learning strategy. Multilayer perceptron (MLP) network is implemented on image datasets to extract pattern modes, which are further realized as electromagnetic pattern masks through the information metasurface. The proposed real‐time imaging system accomplishes classification and imaging tasks with notable performance improvement under noisy environment.
Intelligent solar tracking systems to track the trajectory of the sun across the sky has been actively studied and proposed nowadays. Several low performance intelligent solar tracking systems have ...been designed and implemented. Multilayer perceptron (MLP) is one of the common controllers that used to drive solar tracking systems. However, when the input data are complex for neural network, neural network would not well explain the relationship between these data. Thus, it performed worse than when the input data are simple. Using a premapping of relationship between samples of data as input to neural network along with the original input data could probably a strong guide to help neural network to reach the desired goal and predict the output variables faster and more accurate. It is found that using the output of logistic regression as input to neural network would faster the process of finding the predicted output by neural network. Thus, this study aims to propose new efficient and low complexity single and dual axis solar tracking systems by integrating supervised logistic regression (LR) and supervised MLP or cascade multilayer perceptron (CMLP). LR models are trained by using one of unsupervised clustering techniques (k‐means, fuzzy c‐means, and hierarchical clustering algorithms). The proposed models were used to predict both tilt and orientation angles by two different data sets (month, day, and time variables data set) and (month, day, time, Isc, Voc, and power radiation variables data sets). The results revealed that the proposed MLP/CMLP‐LR systems are able to increase the prediction rate and decrease the mean square error rate as compared to conventional models in both single and dual axis solar tracking systems. The new developed intelligent systems achieved less number of overall connections, less number of neurons, and less time complexity. The finding suggests that the proposed intelligent solar tracking systems has a great potential to be applied for real‐world applications (i.e., solar heating systems, solar lightening systems, factories, and solar powered ventilation.
The contextual-based convolutional neural network (CNN) with deep architecture and pixel-based multilayer perceptron (MLP) with shallow structure are well-recognized neural network algorithms, ...representing the state-of-the-art deep learning method and the classical non-parametric machine learning approach, respectively. The two algorithms, which have very different behaviours, were integrated in a concise and effective way using a rule-based decision fusion approach for the classification of very fine spatial resolution (VFSR) remotely sensed imagery. The decision fusion rules, designed primarily based on the classification confidence of the CNN, reflect the generally complementary patterns of the individual classifiers. In consequence, the proposed ensemble classifier MLP-CNN harvests the complementary results acquired from the CNN based on deep spatial feature representation and from the MLP based on spectral discrimination. Meanwhile, limitations of the CNN due to the adoption of convolutional filters such as the uncertainty in object boundary partition and loss of useful fine spatial resolution detail were compensated. The effectiveness of the ensemble MLP-CNN classifier was tested in both urban and rural areas using aerial photography together with an additional satellite sensor dataset. The MLP-CNN classifier achieved promising performance, consistently outperforming the pixel-based MLP, spectral and textural-based MLP, and the contextual-based CNN in terms of classification accuracy. This research paves the way to effectively address the complicated problem of VFSR image classification.
•Solving cold start problem in cross domain recommender systems.•Trust Aware Cross Domain Deep Neural Matrix Factorization model for rating prediction.•Modeling of linear and non-linear features ...using deep learning.•Incorporating implicit trust relation as trust degree via Ant Colony Optimization.•Evaluating the proposed model on ecommerce dataset that is AliExpress.
Over the years, cross-domain and trust-based recommendation systems are proven to be very helpful in solving issues pertaining to data sparsity and cold start. Many e-commerce sites used recommender systems as business tools for increasing their sale productivity and help their customers in finding suitable products. However, due to sparse rating and lack of historical information, such systems cannot generate effective recommendations. Matrix factorization and deep learning techniques have been the focus of research community for the last few years to solve the problem of data sparsity and cold start. In this paper, we have proposed a model called Trust Aware Cross-Domain Deep Neural Matrix Factorization (TCrossDNMF) that predicts rating of an item for an active user and solves user cold start problem in the cross-domain scenario of ‘User Overlap’ in e-commerce system. TCrossDNMF model is divided into four main steps: i) Features learning that learns the users’ features using a latent factor model and then finds the similarity between users of source and target domains. As the users are shared between two domains, the proposed model learns the common information and transfers the knowledge from a source to target domain. ii) Ranking that finds set of similar users (neighbors), and then filters out the dissimilar users based on similarity threshold θ, and then generates a bipartite trust graph from these reduced set of users and executes Ant Colony Optimization, to find trustworthy neighbors for an active user. iii) Weighting computes the trust degree between an active user and his or her top-k neighbors. iv) Prediction that trains the TCrossDNMF model using multilayer perceptron (MLP) and generalized matrix factorization (GMF) by representing user-item interactions in higher dimensions and ensembles the GMF and MLP with trust information for rating prediction. We evaluated proposed model on a real dataset collected from a popular e-commerce retail service ‘AliExpress’. We used categories available in ‘AliExpress’ as source domain and a target domain. For observing the performance of proposed model, we took six domains that have a higher ratio of sparsity. The proposed model is evaluated by using MAE, RMSE and F-measure metrics and compared it with baselines. The experiments show that the proposed model is a viable solution for the mentioned problem with significant improvements in results.
Cartilage loss due to osteoarthritis (OA) in the patellofemoral joint provokes pain, stiffness, and restriction of joint motion, which strongly reduces quality of life. Early diagnosis is essential ...for prolonging painless joint function. Vibroarthrography (VAG) has been proposed in the literature as a safe, noninvasive, and reproducible tool for cartilage evaluation. Until now, however, there have been no strict protocols for VAG acquisition especially in regard to differences between the patellofemoral and tibiofemoral joints. The purpose of this study was to evaluate the proposed examination and acquisition protocol for the patellofemoral joint, as well as to determine the optimal examination protocol to obtain the best diagnostic results. Thirty-four patients scheduled for knee surgery due to cartilage lesions were enrolled in the study and compared with 33 healthy individuals in the control group. VAG acquisition was performed prior to surgery, and cartilage status was evaluated during the surgery as a reference point. Both closed (CKC) and open (OKC) kinetic chains were assessed during VAG. The selection of the optimal signal measures was performed using a neighborhood component analysis (NCA) algorithm. The classification was performed using multilayer perceptron (MLP) and radial basis function (RBF) neural networks. The classification using artificial neural networks was performed for three variants: I. open kinetic chain, II. closed kinetic chain, and III. open and closed kinetic chain. The highest diagnostic accuracy was obtained for variants I and II for the RBF 9-35-2 and MLP 10-16-2 networks, respectively, achieving a classification accuracy of 98.53, a sensitivity of 0.958, and a specificity of 1. For variant III, a diagnostic accuracy of 97.79 was obtained with a sensitivity and specificity of 0.978 for MLP 8-3-2. This indicates a possible simplification of the examination protocol to single kinetic chain analyses.
Dissolved gas analysis (DGA) of insulating oil in power transformers can offer valuable information related to faults. Due to the poor and unbalanced characteristics of typical DGA datasets, which ...threaten the generalization capability of artificial intelligent (AI)-based models, we propose the use of a new over-sampling technique called ASMOTE (adaptive synthetic minority over-sampling TEchnique) in the pre-processing step to enrich the dataset. ASMOTE can significantly improve the generalization performance of AI-based models by providing a sufficient synthetic dataset to train an AI classifier. To authenticate the effectiveness of the ASMOTE algorithm, we validate the transformer diagnostic accuracy of some typical classification algorithms such as multilayer perceptron (MLP), support vector machine (SVM), and k-nearest neighbor (k-NN) using synthetic datasets created by the SMOTE technique. In addition, the use of DGA ratios is also considered. By investigating the interactions between byproduct gases in insulating oil and transformer faults, the non-code ratios of the dissolved emissions are chosen as the characterizing input to the AI-based models. Moreover, with the ability to extract discriminate faulty information of a transformer from DGA data, MLP is used as a preferable classifier for diagnosing symptoms present in transformers. The empirical results of this study demonstrate that the proposed technique remarkably increases the diagnostic performance of power transformer faults.
Traditional learning algorithms applied to complex and highly imbalanced training sets may not give satisfactory results when distinguishing between examples of the classes. The tendency is to yield ...classification models that are biased towards the overrepresented (majority) class. This paper investigates this class imbalance problem in the context of multilayer perceptron (MLP) neural networks. The consequences of the equal cost (loss) assumption on imbalanced data are formally discussed from a statistical learning theory point of view. A new cost-sensitive algorithm (CSMLP) is presented to improve the discrimination ability of (two-class) MLPs. The CSMLP formulation is based on a joint objective function that uses a single cost parameter to distinguish the importance of class errors. The learning rule extends the Levenberg-Marquadt's rule, ensuring the computational efficiency of the algorithm. In addition, it is theoretically demonstrated that the incorporation of prior information via the cost parameter may lead to balanced decision boundaries in the feature space. Based on the statistical analysis of results on real data, our approach shows a significant improvement of the area under the receiver operating characteristic curve and G -mean measures of regular MLPs.
Despite the successful application of convolutional neural networks (CNNs) in object detection tasks, their efficiency in detecting faults from freight train images remains inadequate for ...implementation in real-world engineering scenarios. Existing modeling shortcomings of spatial invariance and pooling layers in conventional CNNs often ignore the neglect of crucial global information, resulting in error localization for fault objection tasks of freight trains. To solve these problems, we design a spatial-wise dynamic distillation framework based on multilayer perceptron (MLP) for visual fault detection of freight trains. We initially present the axial shift strategy, which allows the MLP-like architecture to overcome the challenge of spatial invariance and effectively incorporate both local and global cues. We propose a dynamic distillation method without a pretraining teacher, including a dynamic teacher mechanism that can effectively eliminate the semantic discrepancy with the student model. Such an approach mines more abundant details from lower level feature appearances and higher level label semantics as the extra supervision signal, which utilizes efficient instance embedding to model the global spatial and semantic information. In addition, the proposed dynamic teacher can jointly train with students to further enhance the distillation efficiency. Extensive experiments executed on six typical fault datasets reveal that our approach outperforms the current state-of-the-art detectors and achieves the highest accuracy with real-time detection at a lower computational cost.