In this paper, we propose to combine an efficient image representation based on local descriptors with a Support Vector Machine classifier in order to perform object categorization. For this purpose, we apply kernels defined on sets of vectors. After testing different combinations of kernel / local descriptors, we have been able to identify a very performant one.

A support vector machine is a new learning machine; it is based on the statistics learning theory and attracts the attention of all researchers. Recently, the support vector machines SVMs have been applied to the problem of financial early-warning prediction Rose, 1999 . The SVMs-based method has been compared with other statistical methods and has shown good results. But the parameters of the ...

This paper proposes a paradigm where commonly made segmental pronunciation errors are modeled as pair-wise confusions between two or more phonemes in the language that is being learnt. The method uses an ensemble of support vector machine classifiers with time varying Mel frequency cepstral features to distinguish between several pairs of phonemes. These classifiers are then applied to classify...

We herein present FactorsR, an RWizard application which provides tools for the identification of the most likely causal factors significantly correlated with species richness, and for depicting on a map the species richness predicted by a Support Vector Machine (SVM) model. As a demonstration of FactorsR, we used an assessment using a database incorporating all species of terrestrial carnivore...

As Internet grows quickly, pornography, which is often printed into a small quantity of publication in the past, becomes one of the highly distributed information over Internet. However, pornography may be harmful to children, and may affect the efficiency of workers. In this paper, we design an easy scheme for detecting pornography. We exploit primitive information from pornography and use thi...

A support vector machine constructs an optimal hyperplane from a small set of samples near the boundary. This makes it sensitive to these specific samples and tends to result in machines either too complex with poor generalization ability or too imprecise with high training error, depending on the kernel parameters. In this paper, we present an improved version of the method, called editing sup...

Discrete support vector machines (DSVM), recently proposed in [l01 and [ l l ] for binary classification problems, have been shown to outperform other competing approaches on well-known benchmark datasets. Here we address their extension to multicategory classification, by developing a one-against-all framework in which a set of binary discrimination problems are solved by means of DSVM. Comput...

The knowledge of soil type and soil texture is crucial for environmental monitoring purpose and risk assessment. Unfortunately, their mapping using classical techniques is time consuming and costly. We present here a way to estimate soil types based on limited field observations and remote sensing data. Due to the fact that the relation between the soil types and the considered attributes that ...

We propose a novel online monitoring approach to distinguish between attacks and normal activity in SIP-based Voice over IP environments. We demonstrate the e ciency of the approach even when only limited data sets are used in learning phase. The solution builds on the monitoring of a set of 38 features in VoIP ows and uses Support Vector Machines for classi cation. We validate our proposal thr...

Article history: Received 20 August 2007 Accepted 27 February 2008 Available online 15 May 2008