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

Znaleziono wyników: 49

Liczba wyników na stronie
first rewind previous Strona / 3 next fast forward last
Wyniki wyszukiwania
Wyszukiwano:
w słowach kluczowych:  support vector machine
help Sortuj według:

help Ogranicz wyniki do:
first rewind previous Strona / 3 next fast forward last
EN
In this paper, we compare the following machine learning methods as classifiers for sentiment analysis: k – nearest neighbours (kNN), artificial neural network (ANN), support vector machine (SVM), random forest. We used a dataset containing 5,000 movie reviews in which 2,500 were marked as positive and 2,500 as negative. We chose 5,189 words which have an influence on sentence sentiment. The dataset was prepared using a term document matrix (TDM) and classical multidimensional scaling (MDS). This is the first time that TDM and MDS have been used to choose the characteristics of text in sentiment analysis. In this case, we decided to examine different indicators of the specific classifier, such as kernel type for SVM and neighbour count in kNN. All calculations were performed in the R language, in the program R Studio v 3.5.2. Our work can be reproduced because all of our data sets and source code are public.
PL
Niniejszy artykuł przedstawia proces dostosowania parametrów modelu maszyny wektorów nośnych, który posłuży do zbadania wpływu wartości parametru długości cyklu sygnalizacji świetlnej na jakość ruchu. Badania przeprowadzono z użyciem danych pozyskanych w trakcie przeprowadzonych symulacji w autorskim symulatorze ruchu ulicznego. W artykule przedstawiono i omówiono wyniki poszukiwania optymalnej wartości parametru długości cyklu sygnalizacji świetlnej.
EN
This article presents the process of adapting support vector machine model’s parameters used for studying the effect of traffic light cycle length parameter’s value on traffic quality. The survey is carried out using data collected during running simulations in author’s traffic simulator. The article shows results of searching for optimum traffic light cycle length parameter’s value.
EN
Cost estimation, as one of the key processes in construction projects, provides the basis for a number of project-related decisions. This paper presents some results of studies on the application of artificial intelligence and machine learning in cost estimation. The research developed three original models based either on ensembles of neural networks or on support vector machines for the cost prediction of the floor structural frames of buildings. According to the criteria of general metrics (RMSE, MAPE), the three models demonstrate similar predictive performance. MAPE values computed for the training and testing of the three developed models range between 5% and 6%. The accuracy of cost predictions given by the three developed models is acceptable for the cost estimates of the floor structural frames of buildings in the early design stage of the construction project. Analysis of error distribution revealed a degree of superiority for the model based on support vector machines.
EN
Parkinson's disease (PD) is a progressive neurological disorder prevalent in old age. Past studies have shown that speech can be used as an early marker for identification of PD. It affects a number of speech components such as phonation, speech intensity, articulation, and respiration, which alters the speech intelligibility. Speech feature extraction and classification always have been challenging tasks due to the existence of non-stationary and discontinuity in the speech signal. In this study, empirical mode decomposition (EMD) based features are demonstrated to capture the speech characteristics. A new feature, intrinsic mode function cepstral coefficient (IMFCC) is proposed to efficiently represent the characteristics of Parkinson speech. The performances of proposed features are assessed with two different datasets: dataset-1 and dataset-2 each having 20 normal and 25 Parkinson affected peoples. From the results, it is demonstrated that the proposed intrinsic mode function cepstral coefficient feature provides superior classification accuracy in both data-sets. There is a significant increase of 10–20% in accuracy compared to the standard acoustic and Mel-frequency cepstral coefficient (MFCC) features.
EN
Diabetes mellitus (DM) is a multifactorial disease characterized by hyperglycemia. The type 1 and type 2 DM are two different conditions with insulin deficiency and insulin resistance, respectively. It may cause atherosclerosis, stroke, myocardial infarction and other relevant complications. It also features neurological degeneration with autonomic dysfunction to meet metabolic demand. The autonomic balance controls the physiological variables that exhibit nonlinear dynamics. Thus, in current work, nonlinear heart rate variability (HRV) parameters in prognosis of diabetes using artificial neural network (ANN) and support vector machine (SVM) have been demonstrated. The digital lead-I electrocardiogram (ECG) was recorded from male Wister rats of 10–12 week of age and 200 ± 20 gm of weight from control (n = 5) as well as from Streptozotocin induced diabetic rats (n = 5). A total of 526 datasets were computed from the recorded ECG data for evaluating thirteen nonlinear HRV parameters and used for training and testing of ANN. Using these parameters as inputs, the classification accuracy of 86.3% was obtained with an ANN architecture (13:7:1) at learning rate of 0.01. While relatively better accuracy of 90.5% was observed with SVM to differentiate the diabetic and control subjects. The obtained results suggested that nonlinear HRV parameters show distinct changes due to diabetes and hence along with machine learning tools, these can be used for development of noninvasive low-cost real-time prognostic system in predicting diabetes using machine learning techniques.
EN
Recent research on Parkinson disease (PD) detection has shown that vocal disorders are linked to symptoms in 90% of the PD patients at early stages. Thus, there is an interest in applying vocal features to the computer-assisted diagnosis and remote monitoring of patients with PD at early stages. The contribution of this research is an increase of accuracy and a reduction of the number of selected vocal features in PD detection while using the newest and largest public dataset available. Whereas the number of features in this public dataset is 754, the number of selected features for classification ranges from 8 to 20 after using Wrappers feature subset selection. Four classifiers (k nearest neighbor, multi-layer perceptron, support vector machine and random forest) are applied to vocal-based PD detection. The proposed approach shows an accuracy of 94.7%, sensitivity of 98.4%, specificity of 92.68% and precision of 97.22%. The best resulting accuracy is obtained by using a support vector machine and it is higher than the one, which was reported on the first work to use the same dataset. In addition, the corresponding computational complexity is further reduced by selecting no more than 20 features.
EN
This study investigates the properties of the brain electrical activity from different recording regions and physiological states for seizure detection. Neurophysiologists will find the work useful in the timely and accurate detection of epileptic seizures of their patients. We explored the best way to detect meaningful patterns from an epileptic Electroencephalogram (EEG). Signals used in this work are 23.6 s segments of 100 single channel surface EEG recordings collected with the sampling rate of 173.61 Hz. The recorded signals are from five healthy volunteers with eyes closed and eyes open, and intracranial EEG recordings from five epilepsy patients during the seizure-free interval as well as epileptic seizures. Feature engineering was done using; i) feature extraction of each EEG wave in time, frequency and time-frequency domains via Butterworth filter, Fourier Transform and Wavelet Transform respectively and, ii) feature selection with T-test, and Sequential Forward Floating Selection (SFFS). SVM and KNN learning algorithms were applied to classify preprocessed EEG signal. Performance comparison was based on Accuracy, Sensitivity and Specificity. Our experiments showed that SVM has a slight edge over KNN.
EN
Acoustical analysis of snoring provides a new approach for the diagnosis of obstructive sleep apnea hypopnea syndrome (OSAHS). A classification method is presented based on respiratory disorder events to predict the apnea-hypopnea index (AHI) of OSAHS patients. The acoustical features of snoring were extracted from a full night’s recording of 6 OSAHS patients, and regular snoring sounds and snoring sounds related to respiratory disorder events were classified using a support vector machine (SVM) method. The mean recognition rate for simple snoring sounds and snoring sounds related to respiratory disorder events is more than 91.14% by using the grid search, a genetic algorithm and particle swarm optimization methods. The predicted AHI from the present study has a high correlation with the AHI from polysomnography and the correlation coefficient is 0.976. These results demonstrate that the proposed method can classify the snoring sounds of OSAHS patients and can be used to provide guidance for diagnosis of OSAHS.
EN
Particulate matters (PMs) are considered as one of the air pollutants generally associated with poor air quality in both outdoor and indoor environments. The composition, distribution and size of these particles hazardously afect the human health causing cardiovascular health problems, lung dysfunction, respiratory problems, chronic obstructive pulmonary disease and lungs cancer. Classifcation models developed by analyzing mass concentration time series data of atmospheric particulate matter can be used for the prediction of air quality and for issuing warnings to protect the health of the public. In this study, mass concentration time series data of both outdoor and indoor particulates matters PM2.5 (aerodynamics size up to 2.5 μ) and PM10.0 (aerodynamics size up to 10.0 μ) were acquired using Haz-Dust EPAM-5000 from six diferent locations of the Muzafarabad city, Azad Kashmir. The linear and nonlinear approaches were used to extract mass concentration time series features of the indoor and outdoor atmospheric particulates. These features were given as an input to the robust machine learning classifers. The support vector machine (SVM) kernels, ensemble classifers, decision tree and K-nearest neighbors (KNN) are used to classify the indoor and outdoor particulate matter time series. The performance was estimated in terms of area under the curve (AUC), accuracy, true negative rate, true positive rate, negative predictive value and positive predictive value. The highest accuracy (95.8%) was obtained using cubic and coarse Gaussian SVM along with the cosine and cubic KNN, while the highest AUC, i.e., 1.00, is obtained using fne Gaussian and cubic SVM as well as with the cubic and weighted KNN.
EN
One of the most important criteria for selecting coal for a given technology are the ash Fusion temperatures (AFTs). An effective way to regulate the AFTs so that they meet the criteria for a given industrial application is to form blends of different coals. The values of the AFTs in the blends are nonadditive, therefore they can't be calculated using the weighted average of the blend components. On the other hand, direct determination of ATFs values requires many additional time-consuming and expensive laboratory tests. Therefore, it is important to develop a solution that, in addition to the effective prediction of the values of AFTs, will also enable optimal selection of components of the blend in terms of its key parameters. The aim of the work was to develop an algorithm for the selection of the optimal coal blends in terms of AFTs for given industrial applications. This algorithm uses nonlinear classifying model which was built using machine learning method, support vector machine (SVM). To carry out the training samples of Polish hard coals from different mines of the Upper Silesian Coal Basin were used. The accuracy of the developed model is 92.3%. The results indicate the effectiveness of the proposed solution, which can find practical application in the form of an expert system used in the coal industry. The paper presents the concept of developed IT tool which has been tested for a selected case.
EN
The purpose of the work was to predict the selected product parameters of the dry separation process using a pneumatic sorter. From the perspective of application of coal for energy purposes, determination of process parameters of the output as: ash content, moisture content, sulfur content, calorific value is essential. Prediction was carried out using chosen machine learning algorithms that proved to be effective in forecasting output of various technological processes in which the relationships between process parameters are non-linear. The source of data used in the work were experiments of dry separation of coal samples. Multiple linear regression was used as the baseline predictive technique. The results showed that in the case of predicting moisture and sulfur content this technique was sufficient. The more complex machine learning algorithms like support vector machine (SVM) and multilayer perceptron neural network (MPL) were used and analyzed in the case of ash content and calorific value. In addition, k-means clustering technique was applied. The role of cluster analysis was to obtain additional information about coal samples used as feed material. The combination of techniques such as multilayer perceptron neural network (MPL) or support vector machine (SVM) with k-means allowed for the development of a hybrid algorithm. This approach has significantly increased the effectiveness of the predictive models and proved to be a useful tool in the modeling of the coal enrichment process.
PL
Celem pracy było prognozowanie wybranych parametrów produktu procesu suchej separacji za pomocą sortera pneumatycznego. Z punktu widzenia zastosowania węgla do celów energetycznych niezbędne jest określenie parametrów procesowych wydobycia, takich jak: zawartość popiołu, zawartość wilgoci, zawartość siarki czy wartość kaloryczna. Prognozowanie przeprowadzono przy użyciu wybranych algorytmów uczenia maszynowego, które okazały się skuteczne w prognozowaniu wyjścia różnych procesów technologicznych, w których zależności między parametrami procesu są nieliniowe. Źródłem danych wykorzystanych w pracy były eksperymenty procesu suchej separacji węgla. Zastosowano wieloraką regresję liniową jako bazową metodę predykcyjną. Wyniki pokazały, że w przypadku przewidywania zawartości wilgoci i siarki technika ta była wystarczająca. Bardziej złożone algorytmy uczenia maszynowego, takie jak maszyna wektorów nośnych (SVM) i perceptron wielowarstwowy (MLP) zostały wykorzystane i przeanalizowane w przypadku zawartości popiołu i wartości opałowej. Ponadto wdrożono technikę k-średnich. Rolą analizy skupień było uzyskanie dodatkowych informacji na temat próbek węgla będących wejściem procesu. Połączenie technik, takich jak perceptron wielowarstwowy (MLP) lub maszyna wektorów nośnych (SVM) z metodą k-średnich pozwoliło na opracowanie hybrydowego algorytmu. Takie podejście znacznie zwiększyło efektywność modeli predykcyjnych i okazało się użytecznym narzędziem w modelowaniu procesu wzbogacania węgla.
EN
The useful life time of equipment is an important variable related to system prognosis, and its accurate estimation leads to several competitive advantage in industry. In this paper, Remaining Useful Lifetime (RUL) prediction is estimated by Particle Swarm optimized Support Vector Machines (PSO+SVM) considering two possible pre-processing techniques to improve input quality: Empirical Mode Decomposition (EMD) and Wavelet Transforms (WT). Here, EMD and WT coupled with SVM are used to predict RUL of bearing from the IEEE PHM Challenge 2012 big dataset. Specifically, two cases were analyzed: considering the complete vibration dataset and considering truncated vibration dataset. Finally, predictions provided from models applying both pre-processing techniques are compared against results obtained from PSO+SVM without any pre-processing approach. As conclusion, EMD+SVM presented more accurate predictions and outperformed the other models.
PL
Okres użytkowania sprzętu jest ważną zmienną związaną z prognozowaniem pracy systemu, a możliwość jego dokładnej oceny daje zakładom przemysłowym znaczną przewagę konkurencyjną. W tym artykule pozostały czas pracy (Remaining Useful Life, RUL) szacowano za pomocą maszyn wektorów nośnych zoptymalizowanych rojem cząstek (SVM+PSO) z uwzględnieniem dwóch technik przetwarzania wstępnego pozwalających na poprawę jakości danych wejściowych: empirycznej dekompozycji sygnału (Empirical Mode Decomposition, EMD) oraz transformat falkowych (Wavelet Transforms, WT). W niniejszej pracy, EMD i falki w połączeniu z SVM wykorzystano do prognozowania RUL łożyska ze zbioru danych IEEE PHM Challenge 2012 Big Dataset. W szczególności, przeanalizowano dwa przypadki: uwzględniający kompletny zestaw danych o drganiach oraz drugi, biorący pod uwagę okrojoną wersję tego zbioru. Prognozy otrzymane na podstawie modeli, w których zastosowano obie techniki przetwarzania wstępnego porównano z wynikami uzyskanymi za pomocą PSO + SVM bez wstępnego przetwarzania danych. Wyniki pokazały, że model EMD + SVM generował dokładniejsze prognozy i tym samym przewyższał pozostałe badane modele.
EN
Stress is one of the most significant health problems in the 21st century, and should be dealt with due to the costs of primary and secondary cares of stress-associated psychological and psychiatric problems. In this study, the brain network states exposed to stress were monitored based on electroencephalography (EEG) measures extracted by complex network analysis. To this regard, 23 healthy male participants aged 18–28 were exposed to a stress test. EEG data and salivary cortisol level were recorded for three different conditions including before, right after, and 20 min after exposure to stress. Then, synchronization likelihood (SL) was calculated for the set of EEG data to construct complex networks, which are scale reduced datasets acquired from multi-channel signals. These networks with weighted connectivity matrices were constructed based on original EEG data and also by using four different waves of the recorded signals including d, u, a, and b. In addition to these networks with weighted connectivity, networks with binary connectivity matrices were also derived using threshold T. For each constructed network, four measures including transitivity, modularity, characteristic path length, and global efficiency were calculated. To select the sensitive optimal features from the set of the calculated measures, compensation distance evaluation technique (CDET) was applied. Finally, multi-class support vector machine (SVM) was trained in order to classify the brain network states. The results of testing the SVM models showed that the features based on the original EEG, a and b waves have got better performances in monitoring the brain network states.
14
Content available remote P300 based character recognition using sparse autoencoder with ensemble of SVMs
EN
In this study, a brain–computer interface (BCI) system known as P300 speller is used to spell the word or character without any muscle activity. For P300 signal classification, feature extraction is an important step. In this work, deep feature learning techniques based on sparse autoencoder (SAE) and stacked sparse autoencoder (SSAE) are proposed for feature extraction. Deep feature provides the abstract information about the signal. This work proposes fusion of deep features with the temporal features, which provides abstract and temporal information about the EEG signal. These deep feature and temporal feature are partially complement of each other to represent the EEG signal. For classification of the EEG signal, an ensemble of support vector machines (ESVM) is adopted as it helps to reduce the classifiers variability. In classifier ensemble system, the score of individual classifier is not at the same level. To transform these scores into a common level, min–max normalization is proposed prior to combining them. Min-max normalization scales the classifiers' score between 0 and 1. The experiments are conducted on three standard public datasets, dataset IIb of BCI Competition II, dataset II of the BCI Competition III and BNCI Horizon dataset. The experimental results show that the proposed method yields better or comparable performance compared to earlier reported techniques.
EN
The proposed work develops a rapid and automatic method for brain tumour detection and segmentation using multi-sequence magnetic resonance imaging (MRI) datasets available from BraTS. The proposed method consists of three phases: tumourous slice detection, tumour extraction and tumour substructures segmentation. In phase 1, feature blocks and SVM classifier are used to classify the MRI slices into normal or tumourous. Phase 2 contains fuzzy c means (FCM) algorithm to extract the tumour region from slices identified by phase 1. In addition, graphics processing unit (GPU) based FCM method has been implemented for reducing the processing time which is major overhead with FCM processing of MRI volumes. For phase 3, a novel probabilistic local ternary patterns (PLTP) technique is used to segment the tumour substructures based on the probability density value of histogram bins. Quantitative measures such as sensitivity, specificity, accuracy and dice values are used to analyses the performance of the proposed method and compare with state-of-art-methods. As post processing, the tumour volume estimation and 3D visualization were done for analyzing the nature and location of the tumour to the medical experts. Further, the availability of the GPU reduces the processing time up to 18 than serial CPU processing.
EN
An upper limb amputation is a traumatic event that can seriously affect the person's capacity to perform regular tasks and can lead individuals to lose their confidence and autonomy. Prosthetic devices can be controlled via the acquisition and processing of electromyogram signal produced at the muscles fiber from the surface of the body with an array of an electrode placed on the residual limb. This paper presents the feasibility of classifying the different shoulder movements from around shoulder muscles of transhumeral amputees. The performance of a classifier is affected by the variation of Surface Electromyography (sEMG) signals due to the different categories of contraction. To avoid this, the wavelet transform and data transformation method are employed for features extraction from sEMG signals. Afterward five different supervised machine learning techniques viz. Support Vector Machines (SVM), Decision Tree (DT), Random Forest (RF), K-Nearest Neighbor (KNN) and Naïve Bayes (NB) are applied to determine the different classifiers accuracy. An effective combination of wavelet and RF achieves the best performance with a total classification accuracy of 98%.
17
Content available remote Evaluation of filters over different stimulation models in evoked potentials
EN
Filtering is a key process which removes unwanted parts of signals. During signal recording, various forms of noises distort data. Physiological signals are highly noise sensitive and to evaluate them powerful filtering approaches must be applied. The aim of this study is to compare modern filtering approaches on scalp signals. Brain activities were generally examined by brain signals like EEG and evoked potentials (EP). In this study, data were recorded from university students whose age between 18 and 25 years with visual and auditory stimuli. Discrete wavelet transforms, singular spectrum analysis, empirical mode decomposition and discrete Fourier transform based filters were used and compared with raw data on classification performance. Higuchi fractal dimension and entropy features were extracted from EEG; P300 features were extracted from EP signals. Classification was applied with support vector machines. All filtered data gave better scores than raw data. Empirical mode decomposition (EMD) and Fourier-based filter yielded lower results than the discrete wavelet-based filter. Singular spectrum analysis gave the best result at 84.32%. The current study suggests that singular spectrum analysis removes noise from sensitive physiological signals, and EMD requires new mode selection procedures before resynthesizing.
EN
Diabetes mellitus (DM) is one of the most widespread and rapidly growing diseases. With its advancement, DM-related complications are also increasing. We used characteristic features of toe photoplethysmogram for the detection of type-2 DM using support vector machine (SVM). We collected toe PPG signal, from 58 healthy and 83 type-2 DM subjects. From each PPG signal 37 different features were extracted for further classification. To improve the performance of SVM and reduce the noisy data we employed hybrid feature selection technique that reduces the feature set of 37 to 10 on the basis of majority voting. Using 10 selected features set, we gained an accuracy of 97.87%, sensitivity of 98.78% and specificity of 96.61%. Further for the validation of our method we need to do random population test, so that it can be used as a non-invasive screening tool. Photoplethysmogram is an economic, technically easy and completely non-invasive method for both physician and subject. With the high accuracy that we obtained, we hope that our work will help the clinician in screening of diabetes and adopting suitable treatment plan for preventing end organ damage.
EN
Mesial temporal sclerosis (MTS) is the commonest brain abnormalities in patients with intractable epilepsy. Its diagnosis is usually performed by neuroradiologists based on visual inspection of magnetic resonance imaging (MRI) scans, which is a subjective and time-consuming process with inter-observer variability. In order to expedite the identification of MTS, an automated computer-aided method based on brain MRI characteristics is proposed in this paper. It includes brain segmentation and hippocampus extraction followed by calculating features of both hippocampus and its surrounding cerebrospinal fluid. After that, support vector machines are applied to the generated features to identify patients with MTS from those without MTS. The proposed technique is developed and evaluated on a data set comprising 15 normal controls, 18 left and 18 right MTS patients. Experimental results show that subjects are correctly classified using the proposed classifiers with an accuracy of 0.94 for both left and right MTS detection. Overall, the proposed method could identify MTS in brain MR images and show a promising performance, thus showing its potential clinical utility.
EN
Purpose: With the end goal to fulfil stringent structural shape of the component in aeronautics industry, machining of Nimonic-90 super alloy turns out to be exceptionally troublesome and costly by customary procedures, for example, milling, grinding, turning, etc. For that reason, the manufacture and design engineer worked on contactless machining process like EDM and WEDM. Based on previous studies, it has been observed that rare research work has been published pertaining to the use of machine learning in manufacturing. Therefore the current research work proposed the use of SVM, GP and ANN methods to evaluate the WEDM of Nimonic-90. Design/methodology/approach: The experiments have been performed on the WEDM considering five process variables. The Taguchi L 18 mixed type array is used to formulate the experimental plan. The surface roughness is checked by using surface contact profilometre. The evolutionary algorithms like SVM, GP and ANN approaches have been used to evaluate the SR of WEDM of Nimonic-90 super alloy. Findings: The entire models present the significant results for the better prediction of SR peculiarities of WEDM of Nimonic-90 superalloy. The GP PUK kernel model is dominating the entire model. Research limitations/implications: The investigation was carried for the Nimonic-90 super alloy is selected as a work material. Practical implications: The results of this study provide an opportunity to conduct contactless processing superalloy Nimonic-90. At the same time, this contactless process is much cheaper, faster and more accurate. Originality/value: An experimental work has been reported on the WEDM of Udimet-L605 and use of advance machine learning algorithm and optimization approaches like SVM, and GRA is recommended. A study on WEDM of Inconel 625 has been explored and optimized the process using Taguchi coupled with grey relational approach. The applicability of some evolutionary algorithm like random forest, M5P, and SVM also tested to evaluate the WEDM of Udimet-L605.The fuzzy- inference and BP-ANN approached is used to evaluate the WEDM process. The multi-objective optimization using ratio analysis approach has been utilized to evaluate the WEDM of high carbon & chromium steel. But this current research work proposed the use of SVM, GP and ANN methods to evaluate the WEDM of Nimonic-90.
first rewind previous Strona / 3 next fast forward last
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.