Emotion recognition from speech using Fisher's discriminant analysis and Bayesian classifier

High-order Fisher's discriminant analysis

Flooding extent cartography with Landsat TM imagery and regularized kernel Fisher's discriminant analysis

This study illustrates a novel model based FDI method for the common mechanical faults arising during the manufacture of loudspeakers. To overcome the drawbacks of the conventional signal based approaches, the Bayesian classification of impulse responses based on a model based fault symptom database is proposed. The loudspeaker model is estimated via IRES and ARMA techniques. The fault symptom database is constructed with a novel nonlinear loudspeaker model. The performances of Principal Component Analysis (PCA) and Fisher’s Discriminant Analysis (FDA) are compared. The results show the effectiveness of the proposed method. It is also shown that the FDA based classifier performs better than PCA in terms of the accuracy and consistency of the healthy baseline estimation. However, the fault isolation is difficult due to the similarities of fault signatures.

Robustness of multivariate image analysis assessed by resampling techniques and applied to FDG-PET scans of patients with Alzheimer's disease

Observational evidence shows changes to North American weather regime occurrence depending on the strength of the lower-stratospheric polar vortex. However, it is not yet clear how this occurs or to what extent an improved stratospheric forecast would change regime predictions. Here we analyze four North American regimes at 500 hPa, constructed in principal component (PC) space. We consider both the location of the regimes in PC space and the linear regression between each PC and the lower-stratospheric zonal-mean winds, yielding a theory of which regime transitions are likely to occur due to changes in the lower stratosphere. Using a set of OpenIFS simulations, we then test the effect of relaxing the polar stratosphere to ERA-Interim on subseasonal regime predictions. The model start dates are selected based on particularly poor subseasonal regime predictions in the European Centre for Medium-Range Weather Forecasts CY43R3 hindcasts. While the results show only a modest improvement to the number of accurate regime predictions, there is a substantial reduction in Euclidean distance error in PC space. The average movement of the forecasts within PC space is found to be consistent with expectation for moderate-to-large lower-stratospheric zonal wind perturbations. Overall, our results provide a framework for interpreting the stratospheric influence on North American regime behavior. The results can be applied to subseasonal forecasts to understand how stratospheric uncertainty may affect regime predictions, and to diagnose which regime forecast errors are likely to be related to stratospheric errors. Significance Statement Predicting the weather several weeks ahead is a major challenge with large potential benefits to society. The strength of the circulation more than 10 km above the Arctic during winter (i.e., the polar vortex) is one source of predictability. This study investigates how forecast error and uncertainty in the polar vortex can impact predictions of large-scale weather patterns called “regimes” over North America. Through statistical analysis of observations and experiments with a weather forecast model, we develop an understanding of which regime changes are more likely to be due to changes in the polar vortex. The results will help forecasters and researchers understand the contribution of the stratosphere to changes in weather patterns, and in assessing and improving weather forecast models.

Revisiting the Performance of PCA Versus FDA Versus Simple Projection for Image Recognition

The essential task of a Brain-Computer Interface (BCI) is to extract the motor imagery features from Electro-Encephalogram (EEG) signals for classifying the thought process. It is necessary to analyse these obtained signals in both the time domain and frequency domains. It is observed that the combination of multiple algorithms increases the performance of the feature extraction process. This paper identifies combinations that have not been attempted previously and improves the accuracy of the overall process, although other authors implemented different combinations of the techniques. The focus is given more on the feature extraction process and frequency bands, which are user-specific and subject-specific frequency bands. In both time and frequency domains, after analysing EEG signals with the time domain parameter, we select the frequency band and the timing while using the Fisher ratio of the time domain parameter (TDP). We used Fisher discriminant analysis (FDA)-type F-score to simultaneously select the frequency band and time segment for multi-class classification. We extracted subject-specific TDP features from the training trials to train the classifier when optimal time-frequency areas were selected for each subject. In this paper, various methods are explored for obtaining the features, which are Time Domain Parameters (TDP), Fast Fourier Transform (FFT), Principal Component Analysis (PCA), R2, Fast Correlation Based Filter (FCBF), Empirical Mode Decomposition (EMD), and Intrinsic time-scale decomposition (ITD). After the extraction process, PCA is used for dimensionality reduction. An efficient result was obtained with the combination of TDP, FFT, and PCA. We used the multi-class Fisher’s linear discriminant analysis (LDA) as the classifier, which was in line with the FDA-type F-score. It is observed that the combination of feature extraction techniques to the frequency bands that were selected by the Fisher ratio and FDA type F-score along with Fisher’s LDA classifier had higher accuracy than the results obtained other researches. A kappa coefficient accuracy of 0.64 is obtained for the proposed technique. Our method leads to better classification performance when compared to state-of-the-art methods. The novelty of the approach is based on the combination of frequency bands and two feature extraction methods.

Summary form only given. In the paper, the comparison of discriminative capabilities of conventional dimensionality reduction methods and the integration of a dimensionality reduction module and recognizer design with minimum classification error rate are discussed. Conventionally, principal component analysis (PCA) and Fisher's linear discriminant (FLD) are two most popular and widely used dimensionality reduction methods for pattern recognition. However, the objectives of these two methods are quite different. PCA basically tries to faithfully keep the original data representation but FLD tries to separate data distribution of different classes. It therefore seems that FLD can provide more discriminative characteristics to pattern recognition than PCA. However, a completely optimal feature extractor can never be anything but an optimal recognizer. It is only when constraints are placed on the classifier that one can formulate nontrivial problems. We apply a minimum error formulation (MEF) to integrate the design of dimensionality reduction module and pattern recognizer. The experimental results show that such an integration provides very good recognition performance on a data set of hand-written Chinese characters even when the feature number has been significantly reduced.

We introduce two enhanced Fisher linear discriminant (FLD) models (EFM) in order to improve the generalization ability of the standard FLD based classifiers such as Fisherfaces. Similar to Fisherfaces, both EFM models apply first principal component analysis (PCA) for dimensionality reduction before proceeding with FLD type of analysis. EFM-1 implements the dimensionality reduction with the goal to balance between the need that the selected eigenvalues account for most of the spectral energy of the raw data and the requirement that the eigenvalues of the within-class scatter matrix in the reduced PCA subspace are not too small. EFM-2 implements the dimensionality reduction as Fisherfaces do. It proceeds with the whitening of the within-class scatter matrix in the reduced PCA subspace and then chooses a small set of features (corresponding to the eigenvectors of the within-class scatter matrix) so that the smaller trailing eigenvalues are not included in further computation of the between-class scatter matrix. Experimental data using a large set of faces-1,107 images drawn from 369 subjects and including duplicates acquired at a later time under different illumination-from the FERET database shows that the EFM models outperform the standard FLD based methods.

Rapid and accurate identification of foodborne pathogenic bacteria is of great importance because they are often responsible for the majority of serious foodborne illnesses. The confocal Raman microspectroscopy (CRM) is a fast and easy-to-use method known for its effectiveness in detecting and identifying microorganisms. This study demonstrates that CRM combined with chemometrics can serve as a rapid, reliable, and efficient method for the detection and identification of foodborne pathogenic bacteria without any laborious pre-treatments. Six important foodborne pathogenic bacteria including S. flexneri, L. monocytogenes, V. cholerae, S. aureus, S. typhimurium, and C. botulinum were investigated with CRM. These pathogenic bacteria can be differentiated based on several characteristic peaks and peak intensity ratio. Principal component analysis (PCA) was used for investigating the difference of various samples and reducing the dimensionality of the dataset. Performances of some classical classifiers were compared for bacterial detection and identification including decision tree (DT), artificial neural network (ANN), and Fisher's discriminant analysis (FDA). Correct recognition ratio (CRR), area under the receiver operating characteristic curve (ROC), cumulative gains, and lift charts were used to evaluate the performance of models. The impact of different pretreatment methods on the models was explored, and pretreatment methods include Savitzky-Golay algorithm smoothing (SG), standard normal variate (SNV), multivariate scatter correction (MSC), and Savitzky-Golay algorithm 1st Derivative (SG 1st Der). In the DT, ANN, and FDA model, FDA is more robust for overfitting problem and offers the highest accuracy. Most pretreatment methods raised the performance of the models except SNV. The results revealed that CRM coupled with chemometrics offers a powerful tool for the discrimination of foodborne pathogenic bacteria.

Groundwater is considered as one of the most important sources for water supply in Iran. The Fasa Plain in Fars Province, Southern Iran is one of the major areas of wheat production using groundwater for irrigation. A large population also uses local groundwater for drinking purposes. Therefore, in this study, this plain was selected to assess the spatial variability of groundwater quality and also to identify main parameters affecting the water quality using multivariate statistical techniques such as Cluster Analysis (CA), Discriminant Analysis (DA), and Principal Component Analysis (PCA). Water quality data was monitored at 22 different wells, for five years (2009-2014) with 10 water quality parameters. By using cluster analysis, the sampling wells were grouped into two clusters with distinct water qualities at different locations. The Lasso Discriminant Analysis (LDA) technique was used to assess the spatial variability of water quality. Based on the results, all of the variables except sodium absorption ratio (SAR) are effective in the LDA model with all variables affording 92.80% correct assignation to discriminate between the clusters from the primary 10 variables. Principal component (PC) analysis and factor analysis reduced the complex data matrix into two main components, accounting for more than 95.93% of the total variance. The first PC contained the parameters of TH, Ca2+, and Mg2+. Therefore, the first dominant factor was hardness. In the second PC, Cl-, SAR, and Na+ were the dominant parameters, which may indicate salinity. The originally acquired factors illustrate natural (existence of geological formations) and anthropogenic (improper disposal of domestic and agricultural wastes) factors which affect the groundwater quality.

The extent of genetic differentiation among 17 Ethiopian populations (249 individuals) of Phytolacca dodecandra (Endod) sampled along altitudinal gradients that varied from 1600 to 3000 m was investigated using random amplified polymorphic DNA (RAPD). The populations were classified into three altitude groups: lowland (1600–2100 m), central-highland (2101–2500 m) and highland (2500–3000 m). Seventy polymorphic loci scored from 12 RAPD primers, singly or in combination with ecogeographical variables (altitude, longitude, latitude, temperature and rainfall), were used for principal component, discriminant, correlation, and stepwise multiple regression analyses. Principal component analysis (PCA) clearly differentiated lowland and the central-highland populations from those of the highlands independent of their geographical regions. Canonical discriminant analysis separated the lowland plants from those of the highlands with the central-highland plants being intermediate. Classificatory discriminant analysis corrected classification of 92.8% of the 249 plants into their respective three altitude groups. Multiple regression analysis identified a strong association between some RAPDs and altitude, temperature and rainfall, while the variation in most RAPDs was explained by combinations of the different ecogeographical variables. It is hypothesised that the different altitude groups may be (1) chemical and/or physiological ecotypes produced as a result of complex interactions of altitude with climatic and/or edaphic factors, or (2) different in ploidy levels. The significant correlations obtained between population means from some RAPDs and altitude and temperature as well as the strong association of some RAPDs with the ecogeographical variables in the multiple regression analysis suggest that part of the RAPD polymorphism could be adaptive, and responsive to environmental selection.

The number of principal components (PCs) is critical parameter of principal component analysis (PCA) and its selection determines the performance of fault detection. In this paper, we pay attention to the relationship between selection of the number of PCs and sensitivity of fault detection. The fault signal-to-noise ratio (fault SNR) that depends on the number of PCs for a certain fault is presented. It indicates the sensitivity of fault detection. Accordingly, the number of PCs that gives the maximum fault SNR is considered as the optimum principal component. The presented method was applied to detection of the sensor fault and process fault with a prior information. Furthermore, in the process fault simulation, Fisher discriminant analysis (FDA) is applied to obtain the fault direction, showing its superior capability for isolating fault data.

Genetic algorithms combined with discriminant analysis for key variable identification

Data-driven diagenetic facies classification and well-logging identification based on machine learning methods: A case study on Xujiahe tight sandstone in Sichuan Basin