We propose two novel methods for reducing dimension in training polynomial networks. We consider the class of polynomial networks whose output is the weighted sum of a basis of monomials. Our first method for dimension reduction eliminates redundancy in the training process. Using an implicit matrix structure, we derive iterative methods that converge quickly. A second method for dimension redu...

In this paper, we present an approach that unifies sub-space feature extraction and support vector classification for face recognition. Linear discriminant, independent component and principal component analyses are used for dimensionality reduction prior to introducing feature vectors to a support vector machine. The performance of the developed methods in reducing classification error and pro...

In the context of texture classification, this article explores the capacity and the performance of some combinations of feature extraction, linear and nonlinear dimensionality reduction techniques and several kinds of classification methods. The performances are evaluated and compared in term of classification error. In order to test our texture classification protocol, the experiment carried ...

Scale Invariant Feature Transform is a widely used image descriptor, which is distinctive and robust in real-world applications. However, the high dimensionality of this descriptor causes computational inefficiency when there are a large number of points to be processed. This problem has led to several attempts at developing more compact SIFT-like descriptors, which are suitable for faster matc...

Non-Local Means is an image denoising algorithm based on patch similarity. It compares a reference patch with the neighboring patches to find similar patches. Such similar patches participate in the weighted averaging process. Most of the computational time for Non-Local Means is consumed to measure patch similarity. In this thesis, we have proposed an improvement where the image patches are pr...

Cancer classification is one major application of microarray data analysis. Due to the ultra high dimensionality nature of microarray data, data dimension reduction has drawn special attention for such type of data analysis. The currently available data dimension reduction methods are either supervised, where data need to be labeled, or computational complex. In this paper, we proposed to use a...

Feature selection is an important step in designing image classification systems. While many automatic feature selection methods exist, most of them are opaque to their users. We consider that users should be able to gain insight into how observations behave in the feature space, since this may allow the design of better features and the incorporation of domain knowledge. For this purpose, we p...

We describe the design and evaluation of a probabilistic interface for music exploration and casual playlist generation. Predicted subjective features, such as mood and genre, inferred from low-level audio features create a 34dimensional feature space. We use a nonlinear dimensionality reduction algorithm to create 2D music maps of tracks, and augment these with visualisations of probabilistic ...

In this paper, we propose a method to compare and visualize spectrograms in a low dimensional space using manifold learning. This approach is divided in two steps: a data processing and dimensionality reduction stage and a feature extraction and a visualization stage. The procedure is applied on different types of data from a hot rolling process, with the aim to detect chatter. Results obtained...

SUMMARY Feature transformation in automatic text classification (ATC) can lead to better classification performance. Furthermore dimen-sionality reduction is important in ATC. Hence, feature transformation and dimensionality reduction are performed to obtain lower computational costs with improved classification performance. However, feature transformation and dimension reduction techniques hav...