oalib
Search Results: 1 - 10 of 100 matches for " "
All listed articles are free for downloading (OA Articles)
Page 1 /100
Display every page Item
Complex Support Vector Machines for Regression and Quaternary Classification  [PDF]
Pantelis Bouboulis,Sergios Theodoridis,Charalampos Mavroforakis,Leoni Dalla
Computer Science , 2013,
Abstract: The paper presents a new framework for complex Support Vector Regression as well as Support Vector Machines for quaternary classification. The method exploits the notion of widely linear estimation to model the input-out relation for complex-valued data and considers two cases: a) the complex data are split into their real and imaginary parts and a typical real kernel is employed to map the complex data to a complexified feature space and b) a pure complex kernel is used to directly map the data to the induced complex feature space. The recently developed Wirtinger's calculus on complex reproducing kernel Hilbert spaces (RKHS) is employed in order to compute the Lagrangian and derive the dual optimization problem. As one of our major results, we prove that any complex SVM/SVR task is equivalent with solving two real SVM/SVR tasks exploiting a specific real kernel which is generated by the chosen complex kernel. In particular, the case of pure complex kernels leads to the generation of new kernels, which have not been considered before. In the classification case, the proposed framework inherently splits the complex space into four parts. This leads naturally in solving the four class-task (quaternary classification), instead of the typical two classes of the real SVM. In turn, this rationale can be used in a multiclass problem as a split-class scenario based on four classes, as opposed to the one-versus-all method; this can lead to significant computational savings. Experiments demonstrate the effectiveness of the proposed framework for regression and classification tasks that involve complex data.
Support Vector Machines for Brain Tumours Cells Classification  [PDF]
C.M. Bentaouza,M. Benyettou
Journal of Applied Sciences , 2010,
Abstract: This research is a study applied to the supervised classification of brain tumours by a method resulting from the artificial intelligence which is the Support Vector Machines. The artificial intelligence quickly moved these last decades, with the evolution of the cerebral imagery to diagnose certain diseases such as the brain tumours by techniques like magnetic resonance imagery in order to treat this disease by the surgery and microscopy to detect the type and the rank of the tumour. The results obtained by the Support Vector Machines are satisfactory from the point of view of time of learning and convergence, which have in particular tendency to learn data too much, thus providing good performances in generalization. On the other hand the Support Vector Machines give automatically a reliable result.
Bagging Support Vector Machines for Leukemia Classification  [PDF]
Gokmen Zararsiz,Ferhan Elmali,Ahmet Ozturk
International Journal of Computer Science Issues , 2012,
Abstract: Leukemia is one of the most common cancer type, and its diagnosis and classification is becoming increasingly complex and important. Here, we used a gene expression dataset and adapted bagging support vector machines (bSVM) for leukemia classification. bSVM trains each SVM seperately using bootstrap technique, then aggregates the performances of each SVM by majority voting. bSVM showed accuracy between 87.5% - 92.5%, area under ROC curve between 98.0% - 99.2%, F-measure between 90.5% - 92.7% and outperformed single SVM and other classification methods. We also compared our results with other study results which used the same dataset for leukemia classification. Experimental results revealed that bSVM showed the best performance and can be used as a biomarker for the diagnose of leukemia disease.
Probabilistic Classification using Fuzzy Support Vector Machines  [PDF]
Marzieh Parandehgheibi
Computer Science , 2013,
Abstract: In medical applications such as recognizing the type of a tumor as Malignant or Benign, a wrong diagnosis can be devastating. Methods like Fuzzy Support Vector Machines (FSVM) try to reduce the effect of misplaced training points by assigning a lower weight to the outliers. However, there are still uncertain points which are similar to both classes and assigning a class by the given information will cause errors. In this paper, we propose a two-phase classification method which probabilistically assigns the uncertain points to each of the classes. The proposed method is applied to the Breast Cancer Wisconsin (Diagnostic) Dataset which consists of 569 instances in 2 classes of Malignant and Benign. This method assigns certain instances to their appropriate classes with probability of one, and the uncertain instances to each of the classes with associated probabilities. Therefore, based on the degree of uncertainty, doctors can suggest further examinations before making the final diagnosis.
Mammogram Classification Using Support Vector Machines
S. Thamarai Selvi,R. Malmathanraj
International Journal of Soft Computing , 2012,
Abstract: The clustered microcalcifications on X-ray mammogram provides an important cue for early detection of breast cancer. Texture analysis methods can be applied to detect clustered micro calcifications in digitized mammograms. The clustered microcalcifications on X-ray mammogram provides an important cue for early detection of breast cancer. Texture analysis methods can be applied to detect clustered micro calcifications in digitized mammograms. In this study a novel 2 stage method for mammogram segmentation is implemented to facilitate automatic segmentation of micro calcification. The first stage is the Modified combined morphological spectral unsupervised Image segmentation. The first stage includes watershed transform, anisotrophic filtering technique, band pass filtering scheme, gradient synthesisation and Complex Wavelet Transform (CWT) subband extraction. The second stage of the segmentation scheme is the Random walkers segmentation technique. Finally, features are derived from the Ridgelet subbands of the segmented image. The cooccurrence matrix features are also used for classification. This study also implements the Support Vector Machines (SVM) for effective classification of Mammogram into Benign or malignant mammogram. The validation of the classification scheme was performed by using the Receiver Operating Curve (ROC) analysis, the overall sensitivity of the technique measured by the value of Az which was found to be ranging from 0.8-0.928.
Classification of Images Using Support Vector Machines  [PDF]
Gidudu Anthony,Hulley Greg,Marwala Tshilidzi
Computer Science , 2007,
Abstract: Support Vector Machines (SVMs) are a relatively new supervised classification technique to the land cover mapping community. They have their roots in Statistical Learning Theory and have gained prominence because they are robust, accurate and are effective even when using a small training sample. By their nature SVMs are essentially binary classifiers, however, they can be adopted to handle the multiple classification tasks common in remote sensing studies. The two approaches commonly used are the One-Against-One (1A1) and One-Against-All (1AA) techniques. In this paper, these approaches are evaluated in as far as their impact and implication for land cover mapping. The main finding from this research is that whereas the 1AA technique is more predisposed to yielding unclassified and mixed pixels, the resulting classification accuracy is not significantly different from 1A1 approach. It is the authors conclusions that ultimately the choice of technique adopted boils down to personal preference and the uniqueness of the dataset at hand.
Apply Support Vector Machines to the Tampered Images` Detection and Classification
Keh-Jian Ma,Tung-Shou Chen,Chun-Liang Tung,Chih-Wei Lin
Asian Journal of Information Technology , 2012,
Abstract: Support vector machines are an effect tool of data classification and regression, and there are widespread applications, such as image classification, image retrieval, face authentication, data analysis prediction and so on. In this paper, a novel image tamper detection based on support vector machines will be proposed and the tampering types will be discussed. First, we cut the protected image into several non-overlapping 8 by 8 blocks and retrieve three sets of feature values from each block, which contain ten maximal and minimal pixels and the sub-band LL coefficients after one-level discrete wavelet transformation. Next, we start to process the image with white and black tampering and obtain the feature value from the tampered image. Lastly, we use the retrieved image feature value to support vector machines` training, then the vector machines will produce a trained module, and this module can recognize and label all types of the tampered image.
IMPROVING THE CLASSIFICATION ACCURACY USING SUPPORT VECTOR MACHINES (SVMS) WITH NEW KERNEL  [cached]
Ashraf Afifi
Journal of Global Research in Computer Science , 2013,
Abstract: In this paper, we introduce a new kernel function called polynomial radial basis function (PRBF) that could improve the classification accuracy of support vector machines (SVMs). The proposed kernel function combines both Gauss (RBF) and Polynomial (POLY) kernels and is stated in general form. It is shown that the proposed kernel converges faster than the Gauss and Polynomial kernels. The accuracy of the proposed algorithm is compared to algorithms based on both Gaussian and polynomial kernels by application to a variety of non-separable data sets with several attributes. We noted that the proposed kernel gives good classification accuracy in nearly all the data sets, especially those of high dimensions. Keywords: Classification problem, SVMs, kernel functions.
Performance Evaluation for Question Classification by Tree Kernels using Support Vector Machines  [cached]
Muhammad Arifur Rahman
Journal of Computers , 2010, DOI: 10.4304/jcp.5.1.32-39
Abstract: Question answering systems use information retrieval (IR) and information extraction (IE) methods to retrieve documents containing a valid answer. Question classification plays an important role in the question answer frame to reduce the gap between question and answer. This paper presents our research work on automatic question classification through machine learning approaches. We have experimented with machine learning algorithms Support Vector Machines (SVM) using kernel methods. An effective way to integrate syntactic structures for question classification in machine learning algorithms is the use of tree kernel (TK) functions. Here we use SubTree kernel, SubSet Tree kernel with Bag of words and Partial Tree kernels. Trade-off between training error and margin, Costfactor and the decay factor has significant impact when we use SVM for the above mentioned kernel types. The experiments determined the individual impact for Trade-off between training error and margin, Cost-factor and the decay factor and later the combined effect for Trade-off between training error and margin, Cost-factor. For each kernel types depending on these result we also figure out some hyper planes which can maximize the performance. Based on some standard data set outcomes of our experiment for question classification is promising.
A Survey on Potential of the Support Vector Machines in Solving Classification and Regression Problems  [PDF]
Luminita STATE,Catalina COCIANU,Doina FUSARU
Informatica Economica Journal , 2010,
Abstract: Kernel methods and support vector machines have become the most popular learning from examples paradigms. Several areas of application research make use of SVM approaches as for instance hand written character recognition, text categorization, face detection, pharmaceutical data analysis and drug design. Also, adapted SVM’s have been proposed for time series forecasting and in computational neuroscience as a tool for detection of symmetry when eye movement is connected with attention and visual perception. The aim of the paper is to investigate the potential of SVM’s in solving classification and regression tasks as well as to analyze the computational complexity corresponding to different methodologies aiming to solve a series of afferent arising sub-problems.
Page 1 /100
Display every page Item


Home
Copyright © 2008-2017 Open Access Library. All rights reserved.