Ograniczanie wyników
Czasopisma help
Autorzy help
Lata help
Preferencje help
Widoczny [Schowaj] Abstrakt
Liczba wyników

Znaleziono wyników: 75

Liczba wyników na stronie
first rewind previous Strona / 4 next fast forward last
Wyniki wyszukiwania
Wyszukiwano:
w słowach kluczowych:  SVM
help Sortuj według:

help Ogranicz wyniki do:
first rewind previous Strona / 4 next fast forward last
PL
Celem pracy jest ocena jakości modelu oparta na Maszynie Wektorów Nośnych SVM pod kątem jej przydatności w wirtualnym uruchomieniu - do zastosowania na potrzeby wirtualnego bliźniaka. Przedstawione wyniki badań są ściśle skorelowane z Przemysłem 4.0, którego główną ideą jest integracja inteligentnych maszyn, systemów i informatyki. Jednym z celów jest wprowadzenie możliwości elastycznej zmiany asortymentu oraz zmian w systemach produkcyjnych. Wirtualne uruchomienie może zostać użyte do stworzenia modelu symulacyjnego obiektu, na potrzeby szkolenia operatorów. Jednym z działów wirtualnego rozruchu jest cyfrowy bliźniak. Jest to wirtualna reprezentacja instalacji lub urządzenia, czy też maszyny. Dzięki zastosowaniu wirtualnego bliźniaka, możliwe jest odwzorowanie różnych procesów w celu obniżenia kosztów procesu i przyspieszenia procesu testowania. W pracy zaproponowano współczynnik oceny jakości modelu oparty na SVM. Współczynnik ten bierze pod uwagę wiedzę ekspercką oraz metody używane do oceny jakości modelu - Znormalizowany Błąd Średniokwadratowy NRMSE (ang. Normalized Root Mean Square Error) oraz Znormalizowany Maksymalny Błąd ME (ang. Maximum Error). Wspomniane metody są powszechnie stosowane do oceny jakości modelu, jednak dotychczas nie były używane równocześnie. W każdej z metod uwzględniany jest inny aspekt dotyczący modelu. Zaproponowany współczynnik umożliwia podjęcie decyzji, czy dany model może zostać użyty do stworzenia wirtualnego bliźniaka. Takie podejście pozwala na testowanie modeli w sposób automatyczny lub półautomatyczny.
EN
This paper proposes a model quality assessment method based on Support Vector Machine, which can be used to develop a digital twin. This work is strongly connected with Industry 4.0, in which the main idea is to integrate machines, devices, systems, and IT. One of the goals of Industry 4.0 is to introduce flexible assortment changes. Virtual commissioning can be used to create a simulation model of a plant or conduct training for maintenance engineers. One branch of virtual commissioning is a digital twin. The digital twin is a virtual representation of a plant or a device. Thanks to the digital twin, different scenarios can be analyzed to make the testing process less complicated and less time-consuming. The goal of this work is to propose a coefficient that will take into account expert knowledge and methods used for model quality assessment (such as Normalized Root Mean Square Error - NRMSE, Maximum Error - ME). NRMSE and ME methods are commonly used for this purpose, but they have not been used simultaneously so far. Each of them takes into consideration another aspect of a model. The coefficient allows deciding whether the model can be used for digital twin appliances. Such an attitude introduces the ability to test models automatically or in a semi-automatic way.
EN
Land cover mapping of marshland areas from satellite images data is not a simple process, due to the similarity of the spectral characteristics of the land cover. This leads to challenges being encountered with some land covers classes, especially in wetlands classes. In this study, satellite images from the Sentinel 2B by ESA (European Space Agency) were used to classify the land cover of Al Hawizeh marsh/Iraq Iran border. Three classification methods were used aimed at comparing their accuracy, using multispectral satellite images with a spatial resolution of 10 m. The classification process was performed using three different algorithms, namely: Maximum Likelihood Classification (MLC), Artificial Neural Networks (ANN), and Support Vector Machine (SVM). The classification algorithms were carried out using ENVI 5.1 software to detect six land cover classes: deep water marsh, shallow water marsh, marsh vegetation (aquatic vegetation), urban area (built up area), agriculture area, and barren soil. The results showed that the MLC method applied to Sentinel 2B images provides a higher overall accuracy and the kappa coefficient compared to the ANN and SVM methods. Overall accuracy values for MLC, ANN, and SVM methods were 85.32%, 70.64%, and 77.01% respectively.
EN
Context: Predicting the priority of bug reports is an important activity in software maintenance. Bug priority refers to the order in which a bug or defect should be resolved. A huge number of bug reports are submitted every day. Manual filtering of bug reports and assigning priority to each report is a heavy process, which requires time, resources, and expertise. In many cases mistakes happen when priority is assigned manually, which prevents the developers from finishing their tasks, fixing bugs, and improve the quality. Objective: Bugs are widespread and there is a noticeable increase in the number of bug reports that are submitted by the users and teams’ members with the presence of limited resources, which raises the fact that there is a need for a model that focuses on detecting the priority of bug reports, and allows developers to find the highest priority bug reports. This paper presents a model that focuses on predicting and assigning a priority level (high or low) for each bug report. Method: This model considers a set of factors (indicators) such as component name, summary, assignee, and reporter that possibly affect the priority level of a bug report. The factors are extracted as features from a dataset built using bug reports that are taken from closed-source projects stored in the JIRA bug tracking system, which are used then to train and test the framework. Also, this work presents a tool that helps developers to assign a priority level for the bug report automatically and based on the LSTM’s model prediction. Results: Our experiments consisted of applying a 5-layer deep learning RNN-LSTM neural network and comparing the results with Support Vector Machine (SVM) and K-nearest neighbors (KNN) to predict the priority of bug reports. The performance of the proposed RNN-LSTM model has been analyzed over the JIRA dataset with more than 2000 bug reports. The proposed model has been found 90% accurate in comparison with KNN (74%) and SVM (87%). On average, RNN-LSTM improves the F-measure by 3% compared to SVM and 15.2% compared to KNN. Conclusion: It concluded that LSTM predicts and assigns the priority of the bug more accurately and effectively than the other ML algorithms (KNN and SVM). LSTM significantly improves the average F-measure in comparison to the other classifiers. The study showed that LSTM reported the best performance results based on all performance measures (Accuracy = 0.908, AUC = 0.95, F-measure = 0.892).
EN
We present in this paper a novel DTC-SVM scheme for a Dual Star Induction Motor fed by two 3 levels inverter with NPC structure, this type of inverter has many points of interest in comparison with the conventional 2 levels inverter, such as: sinus wave forms of the output voltage, low THD of voltage and current and low switching frequency. Also, we present in this article an improved direct power control with virtual flux (DPCVF-SVM) for the control of three phase rectifier. It is shown that the DPC-SVM presents several advantages: constant switching frequency, good dynamic response, sinusoidal line currents etc.
EN
The article contains a review of selected classification methods of dermatoscopic images with human skin lesions, taking into account various stages of dermatological disease. The described algorithms are widely used in the diagnosis of skin lesions, such as artificial neural networks (CNN, DCNN), random forests, SVM, kNN classifier, AdaBoost MC and their modifications. The effectiveness, specificity and accuracy of classifications based on the same data sets were also compared and analyzed.
PL
Artykuł zawiera przegląd wybranych metod klasyfikacji obrazów dermatoskopowych zmian skórnych człowieka z uwzględnieniem różnych etapów choroby dermatologicznej. Opisane algorytmy są szeroko wykorzystywane w diagnostyce zmian skórnych, takie jak sztuczne sieci neuronowe (CNN, DCNN), random forests, SVM, klasyfikator kNN, AdaBoost MC i ich modyfikacje. Porównana i przeanalizowana została również skuteczność, specyficznośc i dokładność klasyfikatów w oparciu o te same zestawy danych.
EN
The study examines the efects of urbanization on land surface temperature (LST) in Guwahati, a city in India using satellite data. Landsat images were utilized for LST retrieval, land-use land-cover (LULC) classifcation and the normalized diference built-up index mapping. Surface Energy Balance Algorithms for Land and support vector machine methods were used in the study. Results showed that the city has gone through massive changes in its LULC pattern with a high degree of urbanization during the period 1992–2015. The built-up area (BUA) increased to 87.8 km2 in 2015 from 11.6 km2 in 1992 while vegetation decreased from 143.3 to 76.6 km2 . Open spaces and water bodies decreased from 14.5 to 5 km2 and to 6.6 km2 from 6.7 km2 , respectively. Conversely, an increasing trend of LST was observed. The mean LST which was 18.5 °C in 1992 rose to 29.03 °C in 2015. Linear regression used in quantifying the relationship between urbanization and LST showed a positive relationship between LST and BUAs in the city.
EN
Students' dropout is certainly one of the major problems that afflict educational institutions, the losses caused by the student's abandonment are social, academic and economic waste. The quest for its causes has been subject of work and educational research around the world. Several organizations seek strategic decisions to control the dropout rate. This work's goal is to evaluate the effectiveness of the most used data mining algorithms in the education area. An "in vivo'' controlled experiment was planned and performed to compare the efficacy selected classifiers. The Random Forest and SVM algorithms have stood out in this context, having, statistically similar accuracy (80.36%, 81.18%), precision (80.79%, 80.25%), recall (76.50%, 77.51%) and f-measure (78.86%, 78.81%) averages. The results showed evidence of significant differences between the algorithms, and also showed that, although the SVM had the best metric of accuracy and recall, it results were statistically similar with Random Forest results.
EN
Saturated hydraulic conductivity (Ks) describes the water movement through saturated porous media. The hydraulic conductivity of streambed varies spatially owing to the variations in sediment distribution profiles all along the course of the stream. The artificial intelligence (AI) based spatial modeling schemes were instituted and tested to predict the spatial patterns of streambed hydraulic conductivity. The geographical coordinates (i.e., latitude and longitude) of the sampled locations from where the in situ hydraulic conductivity measurements were determined were used as model inputs to predict streambed Ks over spatial scale using artificial neural network (ANN), adaptive neuro-fuzzy inference system (ANFIS) and support vector machine (SVM) paradigms. The statistical measures computed by using the actual versus predicted streambed Ks values of individual models were comparatively evaluated. The AI-based spatial models provided superior spatial Ks prediction efficiencies with respect to both the strategies/schemes considered. The model efficiencies of spatial modeling scheme 1 (i.e., Strategy 1) were better compared to Strategy 2 due to the incorporation of more number of sampling points for model training. For instance, the SVM model with NSE = 0.941 (Strategy 1) and NSE = 0.895 (Strategy 2) were the best among all the models for 2016 data. Based on the scatter plots and Taylor diagrams plotted, the SVM model predictions were found to be much efficient even though, the ANFIS predictions were less biased. Although ANN and ANFIS models provided a satisfactory level of predictions, the SVM model provided virtuous streambed Ks patterns owing to its inherent capability to adapt to input data that are non-monotone and nonlinearly separable. The tuning of SVM parameters via 3D grid search was responsible for higher efficiencies of SVM models.
PL
W pracy przedstawiono wyniki badań, których celem było utworzenie modelu pozwalającego na określenie odporności istniejących obiektów mostowych na wpływy wstrząsów górniczych. Podstawą do analiz była utworzona przez autora baza danych o odporności dynamicznej żelbetowych obiektów mostowych poddanych wymuszeniu sejsmicznemu charakterystycznemu dla terenu Legnicko-Głogowskiego Okręgu Miedziowego (LGOM). Odporność dynamiczna każdego obiektu w bazie danych została wyrażona w postaci granicznych wartości przyspieszeń drgań gruntu, jakie dana konstrukcja może przejąć bez zagrożenia bezpieczeństwa. Badania przeprowadzono, wykorzystując metodę Support Vector Machine (SVM) w ujęciu regresyjnym (SVR – Support Vector Regression) oraz sztuczne sieci neuronowe (ANN – Artificial Neural Network). Utworzone w ten sposób modele porównano w aspekcie jakości predykcji oraz uogólniania nabytej wiedzy. Pozwoliło to na wytypowanie metody najbardziej efektywnej pod względem oceny odporności dynamicznej istniejących obiektów mostów.
EN
This paper demonstrates the results of research studies aimed at creating a model that allows to determine the resistance of existing bridge structures to the impact of mining tremors. A database (created by the author of this article) of the dynamic resistance of reinforced concrete bridge structures subjected to seismic excitations commonly occurring in the Legnica-Głogów Copper District (LGOM) formed the basis for the analysis. The dynamic resistance of each structure contained in the database was expressed as the limit values of the acceleration of ground vibrations that may be carried by a given structure without compromising its safety. The study was carried out using the Support Vector Machine (SVM) method in a Support Vector Regression (SVR) approach as well as an Artificial Neural Network (ANN). The models were compared in terms of the quality of the predictions and generalization of the acquired knowledge. This allows to select the most-effective method in evaluating the dynamic resistance of existing bridge structures.
EN
The production capacity of installed wind power greatly increases in worldwide. Hence the interest is focused on the reliability and efficiency of wind turbines; then to reduce the production cost and increase the yield. The main objective of our research in this work is to diagnose wind system. We presented a state of the art of diagnosis approach applied on wind turbines and various occurred faults which should be detected and isolated in the wind turbine parts. After that, an overview on this proposed solution for wind turbines, which opted for a diagnostic strategy based on support vector machines (SVM). A Benchmark of a wind power of 4.5 MW with faults on sensors, actuators and the systems was presented. Defects of the Benchmark are in the pitch system, the drive system, the generator and the converter. We tested then the effectiveness of the used method by visualizing simulation results of diagnosis in two different scenarios.
EN
For the purposes of planning and operation of maritime activities, information about wave height dynamics is of great importance. In the paper, real-time prediction of significant wave heights for the following 0.5–5.5 h is provided, using information from 3 or more time points. In the first stage, predictions are made by varying the quantity of significant wave heights from previous time points and various ways of using data are discussed. Afterwards, in the best model, according to the criteria of practicality and accuracy, the influence of wind is taken into account. Predictions are made using two machine learning methods – artificial neural networks (ANN) and support vector machine (SVM). The models were built using the built-in functions of software Weka, developed by Waikato University, New Zealand.
12
Content available Intelligent touch based user authentication
EN
Many researches had shown that touch based authentication is something possible to implement in many devices. This research focuses mainly on making a progress in this field by using more advanced methods such as SVM, kNN, kmeans or neural networks in attempt to build system for both recognizing and learning user’s behavior.
EN
In this paper the results of automated, vision based classification of oak seeds viability i.e. their ability to germinate are presented. In the first stage, using a photo of the seed cross-section, a set of feature vectors were determined. Then three classification methods were examined: k-nearest neighbours (k-NNs), artificial neural networks (ANNs) and support vector machines (SVMs). Finally, a 73.1% precision was obtained for kNN and a 64 bin histogram, 78.5% for ANN and a 4 bin histogram and 78.8% for SVM with a 64 bin histogram.
PL
W artykule zaprezentowano wyniki badań automatycznej, wizyjnej klasyfikacji nasion dębu pod względem ich żywotności, tj. zdolności do kiełkowania. W pierwszym etapie prac, na podstawie zdjęcia przekroju nasiona, wyznaczono zbiór cech, który w sposób niezależny od kształtu i rozmiaru poszczególnych obiektów pozwala na opisanie ich budowy anatomicznej. Następnie zbadano, dla wyselekcjonowanych wektorów cech, trzy metody klasyfikacji: k-najbliższych sąsiadów (k-NN), artificial neural networks (ANN) oraz maszynę wektorów nośnych (SVM). Uzyskano 73,1% precyzji rozpoznawania dla histogramu o długości 64 metodą kNN, 78,5% dla histogramu o długości 4 dla ANN i 78,8% dla histogramu o długości 64 metodą SVM.
EN
This paper describes a number of experiments to compare and validate the performance of machine learning classifiers. Creating machine learning models for data with wide varieties has huge applications in predictive modelling across multiple domain of science. This work reviews state of the art techniques in machine learning classifiers methods with several extent of magnitude in statistics and key findings that will be helpful in establishing best methodological practices for class predictions. Comprehensive comparative review analysis with statistical validations for various machine learning algorithm for SVM, Bagging, Boosting, Decision Trees and Nearest Neighborhood algorithm on multiple data sets is carried out. Focus on the statistical analysis of the results using Friedman-Test and Wilcoxon Test as well as other interpretative metrics like classification rate, ROC, F-measure are evaluated to benchmark results.
15
Content available Support vector machine in gender recognition
EN
In the paper, Support Vector Machine (SVM) methods are discussed. The SVM algorithm is a very strong classification tool. Its capability in gender recognition in comparison with the other methods is presented here. Different sets of face features derived from the frontal facial image such as eye corners, nostrils, mouth corners etc. are taken into account. The efficiency of different sets of facial features in gender recognition using SVM method is examined.
PL
W artykule przedstawiono możliwość detekcji uszkodzeń węzłów na podstawie analizy proporcji pomiędzy wytypowanymi fragmentami funkcji przejścia (FRF). W ramach zadania wykonano eksperyment na modelu laboratoryjnym dwukondygnacyjnej ramy portalowej, którą poddano testom dynamicznym i dla której określono model modalny. Funkcję przejścia odpowiadającą wybranym punktom układu potraktowano jako sygnał w dziedzinie częstotliwości. Wyznaczono odcięte środków ciężkości kwadratów sygnału wybranych fragmentów funkcji, które następnie potraktowano jako dane wejściowe w metodzie wektorów nośnych. Zaproponowane podejście umożliwia skuteczną detekcję uszkodzeń węzłów badanego modelu.
EN
The article presents the possibility of nodes failures detecting based on the analysis of the proportions between the selected intervals of FRF function. Within the scope of the task an experiment was performed on the laboratory model of a two-storey portal frame, which was subjected to dynamic tests and for which a modal model was defined. FRF function for selected system points was treated as a signal in the frequency domain. For the relevant fragments, the centers of gravity of the signal squares were determined, which were then used as input data in the Support Vector Machines (SVM) method. The proposed approach enables effective detection of connection damage in the tested structure.
EN
The paper presents the results of the analysis of the extent of damage to building structures subjected to mining impacts in the form of tremors and continuous surface deformation. The two methods which were used included the multiple regression analysis and the Support Vector Machine – SVM, which belongs to the socalled Machine Learning. The study used the database of the design, technical condition and potential causes of damage to 199 non-renovated buildings, up to the age of 20 years, of a traditional brick construction, located in the mining area of Legnica-Głogów Copper District (LGOM). The conducted analysis allowed for the qualitative assessment of the influence of mining impacts on the extent of damage to the studied buildings.
PL
W referacie przedstawiono wyniki analizy zakresu uszkodzeń budynków poddanych oddziaływaniom górniczym w postaci wstrząsów oraz ciągłych deformacji terenu. Posłużono się statystyczną metodą regresji wielorakiej oraz metodą wektorów podpierających (Support Vector Machine – SVM) zaliczaną do tzw. uczenia maszynowego (Machine Learning). W badaniach wykorzystano bazę danych o konstrukcji, stanie technicznym i potencjalnych przyczynach uszkodzeń 199 nieremontowanych budynków w wieku do 20 lat, o tradycyjnej konstrukcji murowanej, usytuowanych na terenie górniczym Legnicko-Głogowskiego Okręgu Miedziowego (LGOM). Przeprowadzona analiza pozwoliła na jakościową ocenę wpływu oddziaływań górniczych na zakres uszkodzeń badanych budynków.
EN
The paper presents an improved system to recognition of Fuhrman grading in clear-cell renal carcinoma using an ensemble of classifiers. The novelty of solution includes the segmentation applying wavelet transformation in preprocessing stage, application of few selection methods for feature generation and using the ensemble of classifiers in final recognition step. The wavelet transformation is a very efficient tool for image de-noising and enhancing the edges of cell nuclei. The important distinction to other approaches is that diagnostic features of nuclei, based on the texture, geometry, color and histogram, are selected by using few methods, each relying on different mechanism of selection. These different sets of features have enabled creating the ensemble of classifiers based on the support vector machine and random forest, both cooperating with them. Such approach has led to the significant increase of the quality factors in comparison to the best existing results: sensitivity (the average of this solution 94.3% compared to 91.5%) and specificity (the average 98.6% compared to 97.5%.
PL
W publikacji przedstawiono problem osiadań powierzchni spowodowanych przez odwodnienie górotworu, obserwowanych na terenach górniczych. Przedstawiono możliwość prognozowania tych ruchów z wykorzystaniem narzędzi sztucznej inteligencji. Omówiono dwie metody obliczeniowe: wielowarstwową sieć perceptronową oraz metodę wektorów podtrzymujących. Proces uczenia sieci wykonano na zestawie danych reprezentujących jeden z polskich terenów górniczych. Uzyskane wyniki zaprezentowano w postaci wykresów korelacyjnych danych prognozowanych przez sieci oraz oczekiwanych odpowiedzi (dane wysokościowe). Weryfikację poprawności wytrenowania sieci przeprowadzono na próbce danych nieuczestniczących we wcześniejszej procedurze obliczeniowej. Zaprezentowano najlepsze rezultaty z procesu uczenia sieci MLP oraz SVM. W podsumowaniu wskazano możliwości dalszego rozwoju badań w zakresie wykorzystania sztucznej inteligencji w zagadnieniu osiadań odwodnieniowych obserwowanych na terenach górniczych.
EN
This paper presents a phenomenon of surface subsidence caused by dewatering of rock mass observed in mining areas. The possibility of forecasting these movements by the use of artificial intelligence tools was presented, and two calculation methods discussed: Multilayer Perceptron Network (MLP) and the Support Vectors Machines (SVM). The teaching process of the network was performed on the basis of a data set, representing one of the Polish mining areas. Obtained results were presented in the form of correlation graphs of data forecasted by neural networks and expected responses (elevation data). Verification of network training correctness was conducted on a sample of data not involved in the earlier calculation procedure. The best results of the learning process of MLP and SVM networks were presented. The summary indicated the possibility of further development of research in terms of using artificial intelligence in the issue of drainage subsidence observed in mining areas.
20
Content available remote Offline Handwritten Pre-Segmented Character Recognition of Gurmukhi Script
EN
In this paper, we have proposed a feature extraction technique for recognition of segmented handwritten characters of Gurmukhi script. The experiments have been performed with 7000 specimens of segmented offline handwritten Gurmukhi characters collected from 200 different writers. We have considered the set of 35 basic characters of the Gurmukhi script and have proposed the feature extraction technique based on boundary extents of the character image. PCA based feature selection technique has also been implemented in this work to reduce the dimension of data. We have used k-NN, SVM and MLP classifiers. SVM has been used with four different kernels. In this work, we have achieved maximum recognition accuracy of 93.8% for the 35-class problem when SVM with RBF kernel and 5-fold cross validation technique were employed.
first rewind previous Strona / 4 next fast forward last
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.