Search

There are two key challenges when using a linear discriminant analysis in the high-dimensional setting, including singularity of the covariance matrix and difficulty of interpreting the resulting classifier. Although several methods have been proposed to address these problems, they focused only on identifying a parsimonious set of variables maximizing classification accuracy. However, most methods did not consider dependency between variables and efficacy of selected variables appropriately. To address these limitations, here we propose a new approach that directly estimates the sparse discriminant vector without a need of estimating the whole inverse covariance matrix, by formulating a quadratic optimization problem. Furthermore, this approach also allows to integrate external information to guide the structure of covariance matrix. We evaluated the proposed model with simulation studies. We then applied it to the transcriptomic study that aims to identify genomic markers predictive of the response to cancer immunotherapy, where the covariance matrix was constructed based on the prior knowledge available in the pathway database.

similarity measurements are vital in classifier design used for pattern recognition tasks. Representative vector machines (RVMs), which assign the test example according to its nearest representative vector, can be viewed as a framework of typical classifiers such as the nearest neighbor (NN) classifiers, support vector machines (SVMs), sparse representation-based classification (SRC). This paper attempts to employ a variety of similarity measurements to observe the performance of discriminant vector machines (DVMs), which are motivated from RVMs. The experiments conducted on Yale and ORL databases show that L3 norm outperforms the traditional L2 norm. Furthermore, some other similarity measurements are summarized preliminarily. This work is not only a further study of DVMs, but also a worthy perspective to optimize other classifiers.

Distinguishing among the different seismic volcanic patterns is still one of the most important and labor-intensive tasks for volcano monitoring. This task could be lightened and made free from subjective bias by using automatic classification techniques. In this context, a core but often overlooked issue is the choice of an appropriate representation of the data to be classified. Recently, it has been suggested that using a relative representation (i.e. proximities, namely dissimilarities on pairs of objects) instead of an absolute one (i.e. features, namely measurements on single objects) is advantageous to exploit the relational information contained in the dissimilarities to derive highly discriminant vector spaces, where any classifier can be used. According to that motivation, this paper investigates the suitability of a dynamic time warping (DTW) dissimilarity-based vector representation for the classification of seismic patterns. Results show the usefulness of such a representation in the seismic pattern classification scenario, including analyses of potential benefits from recent advances in the dissimilarity-based paradigm such as the proper selection of representation sets and the combination of different dissimilarity representations that might be available for the same data. HighlightsA representation, based on the DTW measure, is proposed for seismic classification.Recent advances of the dissimilarity based representation are investigated for DTW.Experiments with large scope dataset confirm the suitability of the DTW-space.The proposed space, when derived from spectrograms, is the best representation.Selecting small representation sets reduces the number of required DTW comparisons.

Signal peptides recognition by bioinformatics approaches is particularly important for the efficient secretion and production of specific proteins. We concentrate on developing an integrated fuzzy Fisher clustering (IFFC) and designing a novel classifier based on IFFC for predicting secretory proteins. IFFC provides a powerful optimal discriminant vector calculated by fuzzy intra-cluster scatter matrix and fuzzy inter-cluster scatter matrix. Because the training samples and test samples are processed together in IFFC, it is convenient for users to employ their own specific samples of high reliability as training data if necessary. The cross-validation results on some benchmark datasets indicate that the fuzzy Fisher classifier is quite promising for signal peptide prediction.

In this paper, we proposed face recognition based on improved neighborhood maximum margin. The algorithm builds the k-nearest neighbor graph between the data points. Then assigns completely distinct weights for the interclass and intraclass neighbors of a point by fully considering the class information, and the algorithm formula ask the discriminant vector in range space of locality preserving between-class scatter and range space of locality preserving within-class scatter. At last, it finds a liner mapping by maximizing the margin between the interclass and intraclass neighbors of all points, to improve the classification performance in the new subspace. The experiment result on the UMIST face database shows that the method is feasible.

In this paper, a modified Fisher linear discriminant analysis (FLDA) is proposed and aims to not only overcome the rank limitation of FLDA, that is, at most only finding a discriminant vector for 2-class problem based on Fisher discriminant criterion, but also relax singularity of the within-class scatter matrix and finally improves classification performance of FLDA. Experiments on nine publicly available datasets show that the proposed method has better or comparable performance on all the datasets than FLDA.

An efficient unsupervised feature selection method based on unsupervised optimal discriminant vector is developed to find the important features without using class labels. Features are ranked according to the feature importance measurement based on unsupervised optimal discriminant vector in the following steps. First, fuzzy Fisher criterion is adopted as objective function to derive the optimal discriminant vector in unsupervised pattern. Second, the feature importance measurement based on elements of unsupervised optimal discriminant vector is defined to determine the importance of each feature. The features with little importance measurement are removed from the feature subset. Experiments on UCI dataset and fault diagnosis are carried out to show that the proposed method is very efficient and able to deliver reliable results.

We propose a new feature extraction method based on the modified “plus e -take away f” algorithm and discuss an aspect of the optimization method used in sequential feature extraction, by comparing the proposed method with the orthonormal discriminant vector (ODV) method which belongs to a class of sequential feature extraction. It is shown from experimental results that the proposed method is superior to the ODV method in terms of the error probability.

When users rate interesting objects one often gets two-mode data with missing values as result. In the area of recommender systems (automated) collaborative filtering has been used to analyze such kind of two-mode data. Like collaborative filtering (fuzzy) two-mode clustering can be applied to handle so far unknown ratings of users concerning objects of interest. The aim of this paper is to suggest a new algorithm for (fuzzy) two-mode clustering and compare it to collaborative filtering.

According to the principle of minimum description length, the best statistical classifier is the one that minimizes the sum of the complexity of the model and the description length of the training data. This paper focuses on improving the classification rate through correctly classifying the vectors that are misclassified by classifiers. For this purpose, a new tree-structured version of the algorithm, namely tree-structured discriminant vector quantisation, based on the BFOS algorithm. The major problem of the conventional algorithm is overcome by modifying the pdf of the training vectors using the adaptive-boosting algorithm. This new algorithm is implemented on a set of seven textures from the Brodatz data set.

Grosjean, P., Picheral, M., Warembourg, C., and Gorsky, G. 2004. Enumeration,measurement, and identification of net zooplankton samples using the ZOOSCAN digitalimaging system. e ICES Journal of Marine Science, 61: 518e525.Identifying and counting zooplankton are labour-intensive and time-consuming processesthat are still performed manually. However, a new system, known as ZOOSCAN, has beendesigned for counting zooplankton net samples. We describe image-processing and theresults of (semi)-automatic identification of taxa with various machine-learning methods.Each scan contains between 1500 and 2000 individuals !0.5 mm. We used two trainingsets of about 1000 objects each divided into 8 (simplified) and 29 groups (detailed),respectively. The new discriminant vector forest algorithm, which is one of the mostefficient methods, discriminates between the organisms in the detailed training set with anaccuracy of 75% at a speed of 2000 items per second. A supplementary algorithm tagsobjects that the method classified with low accuracy (suspect items), such that they could bechecked by taxonomists. This complementary and interactive semi-automatic processcombines both computer speed and the ability to detect variations in proportions and greylevels with the human skills to discriminate animals on the basis of small details, such aspresence/absence or number of appendages. After this checking process, total accuracyincreases to between 80% and 85%. We discuss the potential of the system as a standard foridentification, enumeration, and size frequency distribution of net-collected zooplankton.

This article reviews the wealth of different pattern recognition methods that have been used for magnetic resonance spectroscopy (MRS) based tumor classification. The methods have in common that the entire MR spectra is used to develop linear and non-linear classifiers. The following issues are addressed: (i) pre-processing, such as normalization and digitization, (ii) extraction of relevant spectral features by multivariate methods, such as principal component analysis, linear discriminant analysis (LDA), and optimal discriminant vector, and (iii) classification by LDA, cluster analysis and artificial neural networks. Different approaches are compared and discussed in view of practical and theoretical considerations.