Support Vector Machine Optimized by Improved Genetic Algorithm

According to the phase space reconstruction theory of nonlinear system, we propose a prediction method of support vector machine based on genetic algorithm. Using the improved autocorrelation method and Grassberger-Procaccia algorithm to determine the time delay and embedding dimension of chaotic signal, the phase space reconstruction is implemented. The penalty coefficient and the kernel function parameter of support vector machine are optimized by genetic algorithm. Combined with support vector machine, single-step prediction model of the chaotic sequence is set up, so we can detect the weak signal in chaos from the prediction error (including the transient signal and periodic signal). Lorenz attractor and the data from the McMaster IPIX radar sea clutter database are used in the simulation. The proposed method can effectively detect the weak target from chaotic signal. When the signal-to-noise ratio is -89.7704 dB in the chaotic noise background, by using the new method the root mean square error can be reduced by two orders of magnitude, reaching 0.00049521, while the conventional support vector machine can reach only 0.049 under the condition of -54.60 dB.

Automatic recognition of skin symptom plays an importance role in the skin diagnosis and treatment. Feature selection is to increase the classification performance of skin symptom. In this paper, the effects of feature selection on the classification of 4-class skin symptoms (chloasma, blackhead, freckled and comedone) are analyzed. Support vector machine (SVM) is employed to construct classifier, the parameters setting is very crucial to the learning results and generalization ability of SVM, and many irrelevant and redundant features degrade the performance of classification, furthermore, the choice of SVM parameters is influenced by the feature subset taken into account and vice versa. So a joint optimization method for selecting features and SVM parameters is proposed. Genetic algorithms, which have a lot of improvements about fitness, crossover and mutation in comparison with simple genetic algorithms, is to achieve a balance between the classification accuracy and the size of the feature subsets selected and have a fast convergence rate. Furthermore, ‘one-against-one’ approach is used to solve multi-class classification problems and a comparison of the model performances is made among the results of the joint optimization and the single optimization of features or SVM parameters in the experiments. The simulation result shows that our approach can find subsets of features for 4-class skin symptoms classification with higher average cross validation accuracy and a higher rate of convergence compared with single optimization of features or SVM parameters.

A genetic algorithm–support vector machine method with parameter optimization for selecting the tag SNPs

Accurate estimation of river flow can have a significant importance in water resources management. In this study, Genetic programming (GP) and Support Vector Machine (SVM) methods were used to forecast daily discharge of Barandoozchay River. The daily discharge data of Barandoozchay River measured at the Dizaj hydrometric station during 2007 to 2011 was used for modeling, which 80% of the data used for training and remaining 20% used for testing of models. The results showed that in the both of considered methods, the models including discharges of one, two and three days ago had higher accuracy in verification step and the accuracy of models decreased with increasing discharge values. Comparing the performance of GP and SVM methods indicated that, however the accuracy of the GP method with the R=0.978 and RMSE=1.66 (m3/s) was slightly more than SVM method with R=0.976 and RMSE=1.80 (m3/s), but the SVM is easier than GP method. Thus, the SVM method can be used as an alternative method in forecasting daily river discharge.

Abstract. The Support Vector Machine (SVM) is a theoretically superior machine learning methodology with great results in classification of remotely sensed datasets. Determination of optimal parameters applied in SVM is still vague to some scientists. In this research, it is suggested to use the Taguchi method to optimize these parameters. The objective of this study was to detect tree crowns on very high resolution (VHR) aerial imagery in Zagros woodlands by SVM optimized by Taguchi method. A 30 ha plot of Persian oak (Quercus persica) coppice trees was selected in Zagros woodlands, Iran. The VHR aerial imagery of the plot with 0.06 m spatial resolution was obtained from National Geographic Organization (NGO), Iran, to extract the crowns of Persian oak trees in this study. The SVM parameters were optimized by Taguchi method and thereafter, the imagery was classified by the SVM with optimal parameters. The results showed that the Taguchi method is a very useful approach to optimize the combination of parameters of SVM. It was also concluded that the SVM method could detect the tree crowns with a KHAT coefficient of 0.961 which showed a great agreement with the observed samples and overall accuracy of 97.7% that showed the accuracy of the final map. Finally, the authors suggest applying this method to optimize the parameters of classification techniques like SVM.

Recent remote sensing literature has shown that support vector machine (SVM) methods generally outperform traditional statistical and neural methods in classification problems involving hyperspectral images. However, there are still open issues that, if suitably addressed, could allow further improvement of their performances in terms of classification accuracy. Two especially critical issues are: 1) the determination of the most appropriate feature subspace where to carry out the classification task and 2) model selection. In this paper, these two issues are addressed through a classification system that optimizes the SVM classifier accuracy for this kind of imagery. This system is based on a genetic optimization framework formulated in such a way as to detect the best discriminative features without requiring the a priori setting of their number by the user and to estimate the best SVM parameters (i.e., regularization and kernel parameters) in a completely automatic way. For these purposes, it exploits fitness criteria intrinsically related to the generalization capabilities of SVM classifiers. In particular, two criteria are explored, namely: 1) the simple support vector count and 2) the radius margin bound. The effectiveness of the proposed classification system in general and of these two criteria in particular is assessed both by simulated and real experiments. In addition, a comparison with classification approaches based on three different feature selection methods is reported, i.e., the steepest ascent (SA) algorithm and two other methods explicitly developed for SVM classifiers, namely: 1) the recursive feature elimination technique and 2) the radius margin bound minimization method

The specifications of the forests in Kalimantan are part of the vegetation type which is characteristic of dry lowland forests that are rich in biodiversity. The Kalimantan forest area is known as an area that has the highest vascular biodiversity in the world. Until 2014, the rate of forest loss as a source of biodiversity was always high every year. The National Polar-Orbiting Suomi Partnership (Suomi NPP) is a new generation of weather satellite imagery from NASA which currently continues to develop application algorithms for environmental monitoring. The support vector machine (SVM) method is a one-class learning machine method with artificial neural network methods that can recognize patterns from input or examples given and also include supervised learning. This study aims to analyze the effect of each parameter on the SVM method and the combination that can produce the highest accuracy, and analyze the ability of the SVM method for making hotspot maps based on the number of forest fire areas in East Kalimantan. The results of the study showed that the Support Vector Machine (SVM) method on the spectral data produced a total of 85%.

Accuracy of data and information on various types of land use and land cover (LULC) is crucial in watershed management planning. The development of remote sensing technology in predicting LULC has been widely used, such as the Support Vector Machine (SVM) and Random Trees (RT) methods. Both methods are part of the model of machine learning and belong to the category of supervised learning. This method can predict, study, and recognize data patterns from the results of training data. The aim of this research is the use of SVM and RT methods for the identification of LULC. Based on the SVM method analysis results, the forest area is 211.61 km2, coffee plantations are 32.92 km2, shrubs are 16.85 km2, built-up/bare land is 3.40 km2, the water body is 3.17 km2, and cloud/shadow is 6.79 km2. While the RT method obtained a forest area of 215.51 km2, coffee plantations covering an area of 29.32 km2, shrubs are covering an area of 16.32 km2, built-up/bare land surrounding an area of 7.22 km2, water body covering an area of 2.62 km2, and cloud/shadow covering an area of 3.75 km2. Overall, the percentage accuracy of the SVM method of 87%, and the RT of 86%. The success rate of modeling SVM (AUC = 0.73) and RT (AUC = 0.75) based on the results of data validation using the area under the receiver operating characteristic curve (AUC) was considered fair good.

In order to obtain the higher classification accuracy in specific categories for the different feature subset, a hierarchical classification algorithm based on Feature Selection is proposed, and is used for synthetic aperture radar (SAR) image classification, and feature selection is achieved by Genetic algorithm. The algorithm has two main characteristics: one is hierarchical classification which consists of many two-class classifier, and the two-class classifier is trained by the optimal feature subset which is selected according to different categories; the second is the classifier of support vector machine SVM (Support Vector Machine); the two is Genetic algorithm which can search out the optimal feature subset and parameters of support vector machine that is most suitable for the category, by unified coding the feature set and the parameters of SVM to constitute the Chromosome. The experiment on the first polarimetric SAR data show that the algorithm can obtain higher classification accuracy rate.

Flood susceptibility mapping using a novel ensemble weights-of-evidence and support vector machine models in GIS

This paper presents a novel fault diagnosis model for oil-immersed power transformers based on dissolved gas analysis. The model is rooted on the theories of rough set and support vector machine. A fitness function based on attribute dependence is developed to identify fault features to improve classification accuracy of transformer fault samples by using Genetic Algorithm. To get improved classification performance, grid search, genetic algorithm and particle swarm optimization are applied to search parameters of support vector machine. Compared with modified Rogers and back propagation neural network, the superiority of the established model is verified.

To resolve the problem of wetland remote sensing image classification, this paper presents an improved classification algorithm. In this algorithm, genetic algorithm (GA) selection, crossover operation is introduced to the standard particle swarm optimization algorithm (PSO) to form a hybrid particle swarm optimization algorithm (GAPSO). The hybrid algorithm can exploit the advantages of the genetic algorithm and particle swarm algorithm respectively to the full to obtain the global optimal parameters of support vector machine (SVM). Thus the wetland remote sensing image can be classified more accurately. Taking Ningxia Shahu wetland remote sensing images as an example, this paper makes a classification of wetland remote sensing images using optimized support vector machine, and the outcome of the experiment shows that this algorithm has better classification effect than that of similar algorithms. DOI : http://dx.doi.org/10.11591/telkomnika.v12i2.4325

Sirekap is a mobile application that was built to help the public monitor and oversee the development of the 2024 elections held in Indonesia. The research aims to apply the Support Vector Machine algorithm in analyzing sentiment about the use of the Sirekap application in 2024. The Support Vector Machine method is used to classify user sentiment into classes, namely positive, negative, and very negative. The amount of data used is 15,000 data sourced from Sirekap application reviews on Google PlayStore, with more detailed research stages including data collection, data preprocessing, data labeling, visualization, word weighting, and testing and analysis. The results show that the Support Vector Machine algorithm provides an accuracy of 88% for the Sirekap 2024 application. These results are expected to help developers to develop further the Sirekap 2024 application in improving the quality of the application and providing better user comfort Based on the results of the sentiment analysis of Sirekap 2024 application users on the Google Play store using the Support Vector Machine (SVM) method, an accuracy rate of 88% was obtained in classifying the sentiment of reviews into positive, negative, and very negative. This shows that the Support Vector Machine method is quite accurate for sentiment analysis of Indonesian text data. Overall, most reviews are very negative with a percentage reaching 76.9%, followed by negative reviews at 12.6%, and the least are positive reviews at 11%.

In this paper, we develop a low-cost portable spectrometer based on Support Vector Machines (SVM) method. The purpose of the spectrometer development is for halal food research project. The portable spectrometer uses various light-emitting diodes (LEDs), light sensor and Arduino microcontroller. The initial research of the spectrometer is testing various oil, such as palm oil, olive oil, sesame oil, soybean oil, and lard. The waveform analysis of each oil can be distinguished by spectroscopy method. But, by using Support Vector Machines (SVM) method the various oil can be distinguished and classified better. The data measurement, classification, and graphical interface result show on the interface of Raspberry Pi based on Python 2.7. Thus, the spectrometer can be used for lard detection.

Hyperspectral remote sensing imagery, due to its rich source of spectral information, provides an efficient tool for land cover classifications in complex geographical areas. However, the high-dimensional space of this imagery poses two important challenges in the classification process: the Hughes phenomena and the existence of relevant and redundant features. The robustness of Support Vector Machines (SVM) in high-dimensional space makes them an efficient tool for classifying hyperspectral imagery. However, optimum SVM parameter determination and optimum feature selection are the two optimization issues that strongly effect SVM performance. Traditional optimization algorithms can discover optimum solutions in a limited search space with one local optimum. Nevertheless, in high-dimensional space traditional optimization algorithms usually get trapped in a local optimum, therefore it is necessary to apply meta-heuristic optimization algorithms to obtain near-global optimum solutions. This study evaluates the potential of Ant Colony Optimization (ACO) for determining SVM parameters and selecting features. Results obtained from AVIRIS and ROSIS hyperspectral datasets demonstrate the superior performance of SVM, achieved by simultaneously optimizing SVM parameters and subsets of the input feature. For comparison, the evaluation is also performed by applying it to other meta-heuristic optimization algorithms such as simulated annealing, tabu search, and genetic algorithm. The results demonstrate a better performance of the ACO-based algorithm in regards to improving the classification accuracy and decreasing the size of selected feature subsets.