Component Analysis Technique Research Articles

ICA analysis of fMRI with real-time constraints: an evaluation of fast detection performance as function of algorithms, parameters and a priori conditions

Independent component analysis (ICA) techniques offer a data-driven possibility to analyze brain functional MRI data in real-time. Typical ICA methods used in functional magnetic resonance imaging (fMRI), however, have been until now mostly developed and optimized for the off-line case in which all data is available. Real-time experiments are ill-posed for ICA in that several constraints are added: limited data, limited analysis time and dynamic changes in the data and computational speed. Previous studies have shown that particular choices of ICA parameters can be used to monitor real-time fMRI (rt-fMRI) brain activation, but it is unknown how other choices would perform. In this rt-fMRI simulation study we investigate and compare the performance of 14 different publicly available ICA algorithms systematically sampling different growing window lengths (WLs), model order (MO) as well as a priori conditions (none, spatial or temporal). Performance is evaluated by computing the spatial and temporal correlation to a target component as well as computation time. Four algorithms are identified as best performing (constrained ICA, fastICA, amuse, and evd), with their corresponding parameter choices. Both spatial and temporal priors are found to provide equal or improved performances in similarity to the target compared with their off-line counterpart, with greatly reduced computation costs. This study suggests parameter choices that can be further investigated in a sliding-window approach for a rt-fMRI experiment.

PCA-based ANN approach to leak classification in the main pipes of VVER-1000

Abstract This paper presents a neural network based fault diagnosing approach which allows dynamic crack and leaks fault identification. The method utilizes the Principal Component Analysis (PCA) technique to reduce the problem dimension. Such a dimension reduction approach leads to faster diagnosing and allows a better graphic presentation of the results. To show the effectiveness of the proposed approach, two methodologies are used to train the neural network (NN). At first, a training matrix composed of 14 variables is used to train a Multilayer Perceptron neural network (MLP) with Resilient Backpropagation (RBP) algorithm. Employing the proposed method, a more accurate and simpler network is designed where the input size is reduced from 14 to 6 variables for training the NN. In short, the application of PCA highly reduces the network topology and allows employing more efficient training algorithms. The accuracy, generalization ability, and reliability of the designed networks are verified using 10 simulated events data from a VVER-1000 simulation using DINAMIKA-97 code. Noise is added to the data to evaluate the robustness of the method and the method again shows to be effective and powerful.

Component analysis reveals sharp tuning of the local field potential in the guinea pig auditory cortex

Local field potentials (LFPs) recorded in the auditory cortex of mammals are known to reveal weakly selective and often multimodal spectrotemporal receptive fields in contrast to spiking activity. This may in part reflect the wider "listening sphere" of LFPs relative to spikes due to the greater current spread at low than high frequencies. We recorded LFPs and spikes from auditory cortex of guinea pigs using 16-channel electrode arrays. LFPs were processed by a component analysis technique that produces optimally tuned linear combinations of electrode signals. Linear combinations of LFPs were found to have sharply tuned responses, closer to spike-related tuning. The existence of a sharply tuned component implies that a cortical neuron (or group of neurons) capable of forming a linear combination of its inputs has access to that information. Linear combinations of signals from electrode arrays reveal information latent in the subspace spanned by multichannel LFP recordings and are justified by the fact that the observations themselves are linear combinations of neural sources.

Influence of the Number of Principal Components used to the Automatic Speaker Recognition Accuracy

This paper discusses possibilities to reduce dimensionality of the standard MFCC feature vectors by applying the technique of Principal Component Analysis (PCA). The reported experimental results suggest that PCA is an appropriate technique to reduce dimensionality without reducing the accuracy of recognition. The applied automatic speaker recognizer shows that already for a 14-dimensional PCA feature space, the recognition accuracy reaches the target value as in the 39-dimensional MFCC feature space. This gives motivation for further research towards more efficient speaker recognizers. Ill. 3, bibl. 9 (in English; abstracts in English and Lithuanian). DOI: http://dx.doi.org/10.5755/j01.eee.123.7.2379

Effects of soil and water conservation and its interactions with soil properties on soil productivity

The effects of soil and water conservation (SWC) on soil properties are well documented. However, definitive and quantitative information of SWC and its interactions with soil properties on soil productivity is lacking for hilly red soil region of southern China. Experiments were conducted in the hilly red soil region of southern China for seven years in three runoff plots, each of which represented different SWC forest-grass measures. Principal component analysis and multiple regression techniques were used to relate the aboveground biomass (representing soil productivity) to soil properties. Based on the final regression equations, soil organic carbon content (Soc) is significantly correlated with soil productivity under the condition of forest-grass measures, whereas pH value and cation exchange capacity (Cec) are the main factors for soil productivity without SWC. Therefore, SWC plays an important role in sequestering Soc and improving soil productivity.

Selection of soil physical quality indicators in relation to soil erodibility

Soil quality indices based on soil characteristics can be used to assess the sustainability of soil and to assist in soil management decisions. Principal component analysis (PCA) technique was used to identify dominant soil characteristics in relation to soil erodibility in watersheds of submontane Punjab (India). Soil physical and chemical characteristics were evaluated for four locations with four land uses at each location whereas runoff, soil loss and soil erodibility were determined at two locations under natural rainfall conditions and at four locations under simulated rainfall conditions. PCA was performed on 22 physical and chemical soil characteristics, which grouped these soil characteristics into five distinct principal components (PCs). These five PCs namely soil hydraulic factor, density factor, structural factor, sand factor and cation factor, explained 86% variability in data. These PCs also explained 86, 96 and 93% variability under natural rainfall conditions and 75, 76 and 77% variability under simulated rainfall conditions in relation to runoff, soil loss and soil erodibility, respectively. Soil total organic carbon content can be considered as dynamic soil physical quality indicator and can be used to monitor temporal and spatial changes in soil quality.

Determination of Phenolic Compounds in Wines

Wine contains natural antioxidants such as phenolic compounds also known as bioactive compounds. Samples of commercially available Greek wines were analyzed in order to determine this phenolic content. For the analysis, Reversed Phase-High Performance Liquid Chromatography (RP-HPLC) coupled with a multiwavelength Ultraviolet/visible (UV/vis) detector was used. The most abundant phenolic substances detected were (+)-catechin (13.5-72.4 mg L-1 ), gallic acid (0.40-99.47 mg L-1) and caffeic acid (0.87-33.48 mg L-1). The principal component analysis (PCA) technique was used to study differentiation among wines according to their production area. Red wines contained more phenolic substances than white ones. Differences of the phenolic composition in wines of the same cultivar were investigated too.

Molecular Dynamics of HIV1-Integrase in Complex with 93del—A Structural Perspective on the Mechanism of Inhibition

HIV1 integrase is an important target for the antiviral therapy. Guanine-rich quadruplex, such as 93del have been shown to be potent inhibitors of this enzyme and thus representing a new class of antiviral agents. Although X-ray and NMR structures of HIV1 integrase and 93del have been reported, there is no structural information of the complex and the mechanism of inhibition still remains unexplored. A number of computational methods including automated protein-DNA docking, and molecular dynamics simulation in explicit solvent were used to model the binding of 93del to HIV1 integrase. Analysis of the dynamic behaviour of the complex using principal components analysis and elastic network modelling techniques allow us to understand how the binding of 93del aptamer and its interactions with key residues affect the intrinsic motions of the catalytic loops by stabilising them in a catalytically inactive conformations. Such insights into the structural mechanism of inhibition can aid in improving the design of anti-HIV aptamers.

Independent Component Analysis–Based Fuel Type Identification for Coal-Fired Power Plants

Independent component analysis (ICA) and support vector machine (SVM) techniques were used to identify the fuel types. Flame oscillation signals were captured by a flame monitor. Thirty flame features were extracted from each flame oscillation signal to form an original feature vector. The ICA technique was applied to choose the independent flame features from each original feature vector. An SVM model was deployed to map the flame features to an individual type of fuel. The results obtained by using eight different types of coal demonstrated that the ICA technique combining with a well trained SVM can be used for identifying the fuel types, and the average success rate was 96.2% in 20 trials. The ICA preceded by principal component analysis (PCA) used for whitening and dimension-reducing performed a bit better than individually using the ICA technique, and the average success rate of fuel type identification was 97.8% in 20 trials.

AGES AND METALLICITIES OF CLUSTER GALAXIES IN A779 USING MODIFIED STRÖMGREN PHOTOMETRY

In the quest for the formation and evolution of galaxy clusters, Rakos and co-workers introduced a spectrophotometric method using modified Stromgren photometry, but with the considerable debate toward the project's abilities, we re-introduce the system by testing for the repeatability of the modified Stromgren colors and compare them with the Stromgren colors, and check for the reproducibility of the ages and metallicities (using the Principle Component Analysis (PCA) technique and the GALEV models) for the six common galaxies in all three A779 data sets. As a result, a fair agreement between two filter systems was found to produce similar colors (with a precision of 0.09 mag in (uz – vz), 0.02 mag in (bz – yz), and 0.03 mag in (vz – vz)) and the generated ages and metallicities are also similar (with an uncertainty of 0.36 Gyr and 0.04 dex from PCA and 0.44 Gyr and 0.2 dex using the GALEV models). We infer that the technique is able to relieve the age-metallicity degeneracy by separating the age effects from the metallicity effects, but it is still unable to completely eliminate it. We further extend this paper to re-study the evolution of galaxies in the low mass, dynamically poor A779 cluster (as it was not elaborately analyzed by Rakos and co-workers in their previous work) by correlating the luminosity (mass), density, and radial distance with the estimated age, metallicity, and the star formation history. Our results distinctly show the bimodality of the young, low-mass, metal-poor population with a mean age of 6.7 Gyr (± 0.5 Gyr) and the old, high-mass, metal-rich galaxies with a mean age of 9 Gyr (± 0.5 Gyr). The method also observes the color evolution of the blue cluster galaxies to red (Butcher-Oemler phenomenon), and the downsizing phenomenon. Our analysis shows that modified Stromgren photometry is very well suited for studying low- and intermediate-z clusters, as it is capable of observing deeper with better spatial resolution at spectroscopic redshift limits, and the narrowband filters estimate the age and metallicity with fewer uncertainties compared to other methods that study stellar population scenarios.

NDVI-rainfall relationship using hyper-temporal satellite data in a portion of North Central Mexico (2000-2010)

Nowadays, availability of hyper-temporal satellite data offers options for studying natural environments that were unattainable by traditional measurement approaches. This work focuses on monitoring vegetation using the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor and rainfall data in the state of San Luis Potosi, Mexico. A hyper-temporal dataset comprising 489 MODIS MCD43A4 imagery (500 m spatial resolution) taken over the last 11 years (2000 to 2010) was used. Ten temporal yearly composites were constructed by fusing normalized difference vegetation indices (NDVI) data along each year using the ‘maximum composite value approach’. The time-series dataset was also fused using the principal component analysis (PCA) technique; the second principal component captured the main inter-annual anomalies from maximum yearly NDVI values. The analysis of this time-series indicates two periods: a) 2000 to 2005 characterized by a drought during 2000, 2001 and 2005, and b) 2006 to 2010 marked by a trend indicating wetter conditions with maxima in 2008 and 2010. Finally, an upward trend in NDVI time-series was observed during the period studied and successfully validated with ancillary historical data of crop production area. Key words: Moderate resolution imaging spectroradiometer, MCD43A4, time-series, principal component analysis, maximum value composite.

Extraversion and fronto-posterior EEG spectral power gradient: An independent component analysis

Several studies show that the fronto-posterior EEG spectral power gradient is a stable individual characteristic related to personality. Whether this characteristic is specifically related to agentic extraversion and theta band of frequencies or is associated with a broader set of personality traits and frequency bands is a matter of debate, as well as the specific cortical regions contributing to this effect. To clarify these questions, we used group independent component analysis (ICA) and source localization techniques. Agentic extraversion was associated with higher theta activity in the default mode network's (DMN) posterior hub and lower theta activity in the orbitofrontal cortex (OFC). Regression analyses showed that theta activity predicted agentic extraversion better than other frequency bands and agentic extraversion predicted posterior versus frontal activity better than other personality dimensions. These results are taken to indicate higher tonic activity in OFC and lower activity in DMN in extraverts as compared to introverts.

The role of organisational infrastructure in successful ERP implementation: an empirical study by hierarchical regression and PCA

Organisations face considerable challenges in implementing and using ERP systems. Several researchers have tried to study different aspects of this system. In this paper, based on literature review and experts' interviews, a conceptual model is proposed which poses that technical and organisational infrastructures have an intervening role in the relationship between critical success factors and ERP performance. A 40-item questionnaire is used for data collection and analysis. Hierarchical regression results proved that at 95% confidence level, infrastructure variable plays a moderating role in the relationship between critical success factors and ERP performance. Also, principal component analysis (PCA) and latent moderated structural equations (LMSs) technique are used to validate the results of hierarchical regression. Results show that organisational infrastructure has the most important effect on ERP performance and 'strategic goals' is the most important critical success factor. To our knowledge, this is the first study which is concerned with the moderating role of organisational infrastructures in the relationship between ERP critical success factors and performance.

An Improved Pca Fusion Method Based on Generalized Intensity–Hue–Saturation Fusion Technique

Abstract Among various image fusion methods, principal component analysis (PCA) technique is capable of quickly merging the massive volumes of data. For IKONOS imagery, PCA can yield satisfactory “spatial” enhancement but may introduce spectral distortion, appearing as a change in colors between compositions of resembled and fused multi-spectral bands. To solve this problem, a fast improved PCA fusion method based on Intensity–Hue–Saturation Fusion Technique with Spectral Adjustment is presented. The experimental results demonstrate that the proposed approach can provide better performance than the original PCA method both in processing speed and image quality.

Organizational Factors Influencing Performance Management System in Higher Educational Institution of South East Asia

The main purpose of this article is to identify the factors that are influencing the implementation of performance management system (PMS) in South East Asia. This study specifically inspects organizational factors influencing PMS. The study adopts a quantitative approach using questionnaires distributed to 123 academics across a University in Malaysia. Factor analysis was conducted using Principle Component Analysis (PCA) technique to identify the construct validity of factors influencing PMS. The results of the study demonstrated that three organizational factors influenced PMS and its effectiveness at a moderate level.

Statistical Analysis of the Convective Intermittent Transport at the Edge Region of QUEST

Ambient fluctuations and the generation of coherent convective structures are investigated as a function of vertical to toroidal magnetic field ratio (Bz/Bt) in the spherical tokamak QUEST [K. Hanada et al., Plasma Fusion Research 5, S1007]. Fluctuation signals are obtained at a high spatial and temporal resolution using the fast visible camera. The coherent structures (blobs) were seen to be apparent only beyond a critical Bz/Bt ratio (~ 10). This clearly indicates a shift from the drift to the interchange mode with the increase in vertical magnetic field. The blobs are propagated radially at a high speed ~ 10 m/s which is ~ 1/10 of the ion acoustic speed in QUEST. Lifetime of the blobs is about 700 μs. Second part of the study comprises of the investigation of the finer structures in the current ramp-up phase which is characterized by lower levels of fluctuations. Singular value decomposition (SVD) and principal component analysis (PCA) techniques are applied on the fast camera images to reveal the structures which are otherwise embedded in the overall intensity image. Electron banana orbits were seen to form from the slab annular plasma and grow wider to eventually shape the plasma current front.

An innovative face image enhancement based on principle component analysis

In this paper, we propose an innovative face hallucination approach based on principle component analysis (PCA) and residue technique. First, the relationship of projection coefficients between high-resolution and low-resolution images using PCA is investigated. Then based on this analysis, a high resolution global face image is constructed from a low resolution one. Next a high-resolution residue is derived based on the similarity between the projections on high and low resolution residue training sets. Finally by combining the global face and residue in high resolution, a high resolution face image is generated. Also the recursive and two-stage methods are proposed, which improve the results of face image enhancement. Extensive experiments validate the proposed approaches.

Assessing thermalization and estimating the Hamiltonian with output data only

I consider the generic situation where a finite number of identical test systems in varying (possibly unknown) initial states are subjected independently to the same unknown process. I show how one can infer from the output data alone whether the process in question induces thermalization and, if so, which constants of the motion characterize the final equilibrium states. In case thermalization does occur and there is no evidence for constants of the motion other than energy, I further show how the same output data can be used to estimate the test systems' effective Hamiltonian. For both inference tasks I devise a statistical framework inspired by the generic techniques of factor and principal component analysis. I illustrate its use in the simple example of qubits.

Comparison of INAA elemental composition data between Lago Grande and Osvaldo archaeological sites in the central Amazon: a first perspective

In this work, 50 ceramic fragments from the Lago Grande and 30 from the Osvaldo archaeological site were compared to assess elemental similarities. The aim is to perform a preliminary comparison between the sites, which are located in the central Amazon, Brazil. The analytical technique employed to obtain the ceramics elemental composition was instrumental neutron activation analysis (INAA). The data set obtained was explored by the multivariate statistical techniques of cluster, principal component and discriminant analysis. The analyzed elements were: Na, Lu, U, Yb, La, Th, Cr, Cs, Sc, Fe, Eu, Ce and Hf. The results showed the existence of at least two compositional groups for Lago Grande and Osvaldo. Each compositional group of Osvaldo archaeological site matches with one group of Lago Grande. Correlated with the archaeological background, the results suggest commercial or cultural exchange in the region, which is an indicative of socio-cultural interactions between those sites.

Tool path generation for triangular meshes using least-squares conformal map

Tool path generation is an interesting and challenging task in free-form surface machining. Iso-planar tool path generation is one of the common approaches to dealing with this task. However, it suffers from an inherent drawback that intersecting intervals of the iso-planar method are limited to surface geometric features. This results in poor machining efficiency in flat regions due to redundant machining paths. For this problem, a new tool path generation method for triangular meshes is proposed based on the least-squares conformal map (LSCM). After LSCM parameterisation with minimal stretching energy, the triangular meshes are unfolded on a plane, where the principal component analysis (PCA) technique is employed to determine a suitable drive line for calculating cutter contact paths by the iso-planar strategy. Therefore, the tool paths are generated in a plane and the unevenness of the traditional iso-planar method is improved. The feasibility and effectiveness of the developed method is demonstrated through a test experiment.