No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Abstract
Many redundant and irrelevant features not only slow training and classifying, increase computational consumption heavy but also reduce the accuracy of detection classification, especially when coping with big data. The features selection is an important issue for intrusion detection, an intrusion detection model based on SVM mixed with Fisher Criterion and SFS is proposed in this paper. In this method, the weights representing the relative importance of each feature are first obtained by Fisher Criterion. The features were joined to the feature subset one at a time, the ones with higher weight first, we run the classifier. Eventually, the optimized feature subset is obtained when a steady and satisfactory test result occurs. The performance of this algorithm is verified on MATLAB2012 platform in the KDD CUP 99 data, and the experimental results indicate good feasibility of the method.
... In this paper the issues like preprocessing of medical data, reclassification of the training sets and determining the importance of classes, formation of reference tables, selection of an informative features set that differentiate between class objects, formed by medical professionals are discussed and solved. Mainly in the most studied references [5][6][7][8][11][12][13]] the Fisher's criterion is used to obtain solutions to problems/tasks. Also for solving problems, the algorithms for an estimate calculation as well as the related software programs are used. ...
... The first step is to build a reference table, based on the importance of the features and objects as well as their contribution to the classes [1-4, 9, 10]; the second step is concerned with the choice of the most useful characteristic features set to be investigated. This corresponds to solving the issue of selection of set of informative features from a given table, their visualization, and the determination of the contribution of the features set to the formation of classes [1][2][3][4][5][6][7][8][9][10][11][12][13]. ...
... Fisher's Discriminant Ratio (FDR) for the scalar part and Sequential Forward Selection (SFS) for vector selection has been chosen. To quantify the discriminatory ratio of individual features between two classes FDR is employed [19]. As can be seen in Table 2, the parameter with a higher Fisher value is more important in distinguishing features. ...
Purpose: Parkinson's Disease (PD) is a neuro-degenerative interminable issue causing dynamic loss of dopamine-creating synapses, which is one of the most far reaching ailments after Alzheimer's infection. In this paper, a system for the classification of Parkinson’s disease tremor using noninvasive measurement and frequency domain features is represented.
Materials and Methods: Tremor time-series of Parkinson's disease patients were recorded via a smartphone’s accelerometer sensor. Short-Time Fourier Transform (STFT) was applied to transform the time-domain signal into the frequency domain with high time-frequency resolution. Several frequency features, including mean, max of power spectral density and side frequency have been extracted and by using the FDR algorithm combinations of features carried enough information to reliably assess the severity of tremor in Parkinson patients were determined.
Results: Four different classifiers were implemented to estimate the severity of tremors based on the Unified Parkinson's Disease Rating Scale (UPDRS) in Parkinson's disease patients.
Conclusion: Classifiers’ estimation was compared to clinical scores derived via neurologist UPDRS annotation on Parkinson's disease patients’ tremor. The best accuracy achieved was 95.91±1.51.
There are studied Informational methods for analyzing and managing systems under uncertainty. The expediency of the use of information uncertainty in the tasks of identification of control objects and the synthesis of regulatory systems is substantiated. For the numerical evaluation of information uncertainty, the amount of disinformation is used as negative Bongard useful information. Such information uncertainty can serve as a criterion for the adequacy of the mathematical model of the control object. This makes it possible to compare several models and select the most adequate model that contributes the least amount of disinformation, and also provides a check of the statistical hypothesis about the adequacy of a particular model. If the criterion value exceeds a certain critical value, the adequacy hypothesis must be rejected.
To calculate the critical values of the information adequacy criterion, a statistical experiment was performed. Using the Monte Carlo method, the probability distribution of the information criterion was investigated. A sufficiently smooth empirical criterion distribution function was constructed. The distribution of the information criterion has a pronounced asymmetry and a small positive kurtosis. It is revealed that this distribution is best approximated by the Generalized Extreme Value Distribution law. The critical value can be defined as a quantile of the level of 0.01 or 0.05 of this distribution.
Keywords: adequacy, computer modeling, distribution law, identification, information criterion, Monte Carlo method, controlled object, statistical hypothesis, uncertainty.
ResearchGate has not been able to resolve any references for this publication.