Principal Component Weights Research Articles

Towards a multi‐layers anomaly detection framework for analyzing network traffic

SummaryAnomaly detection plays a crucial part in identifying unforeseen attacks for network and information security. However, the accuracy of existing network anomaly detection approaches is limited because of the lack of sufficient and high‐quality features. Most research works only take information from one network layer into account, which leads to a situation that some key features of other network layers are omitted. To address this issue, we propose a novel approach, named Multi‐Layers Anomaly Detection, which extracts and combines features from different network layers. In order to reduce redundancy and noise derived from the combination of multiple layers, an algorithm called RanPF is designed by applying principal components analysis (PCA) into random forest (RF) algorithm. RanPF uses features selected by PCA to decide the height of every tree in RF and provides a method to select which features for tree nodes to use according to the weights of principal components. To obtain high‐quality features, we adopt an attribute learning mechanism. Naive Bayes is used to characterize the attribute information, which is fast and simple compared with other learning algorithms such as SVM. In addition, a series of experiments conducted on two real‐life datasets demonstrate that our approach outperforms the state‐of‐the‐art methods in terms of detection rate and false alarm rate. MLAD achieves about 99% detection rate and about 0.6% false alarm rate on average when the ratio of the training set is 60%. Copyright © 2016 John Wiley & Sons, Ltd.

Men and women have similarly shaped carpometacarpal joint bones

Characterizing the morphology of the carpometacarpal (CMC) joint bones and how they vary across the population is important for understanding the functional anatomy and pathology of the thumb. The purpose of this paper was to develop a statistical shape model of the trapezium and first metacarpal bones to characterize the size and shape of the whole bones across a cohort of 50. We used this shape model to investigate the effects of sex and age on the size and shape of the CMC joint bones and the articulating surface area of the CMC joint. We hypothesized that women have similar shape trapezium and first metacarpal bones compared to men, following scaling for overall size. We also hypothesized that age would be a significant predictor variable for CMC joint bone changes. CT image data and segmented point clouds of 50 CMC bones from healthy adult men and women were obtained from an ongoing study and used to generate two statistical shape models. Statistical analysis of the principal component weights of both models was performed to investigate morphological sex and age differences. We observed sex differences, but were unable to detect any age differences. Between men and women the only difference in morphology of the trapezia and first metacarpal bones was size. These findings confirm our first hypothesis, and suggest that the women have similarly shaped trapezium and first metacarpal bones compared to men. Furthermore, our results reject our second hypothesis, indicating that age is a poor predictor of CMC joint morphology.

중소기업 경기예측 모형 및 지수에 관한 연구

The role of small and medium enterprises as an economic growth factor has been accentuated; consequently, the need to develop a business forecast model and indexes that accurately examine business situation of small and medium enterprises has increased. Most current business model and indexes concerning small and medium enterprises, released by public and private institutions, are based on Business Survey Index (BSI) and depend on subjective (business model and) indexes; therefore, the business model and indexes lack a capacity to grasp an accurate business situation of these enterprises. The business forecast model and indexes suggested in the study have been newly developed with Principal Component Analysis(PCA) and weight method to accurately measure a business situation based on reference dates addressed by the National Statistical Office(NSO). Empirical studies will be presented to prove that the newly proposed business model and indexes have their basis in statistical theory and their trend that resembles the existing Composite Index.

The Role of Visible Wealth for Deprivation

Motivated by the lack of literature linking actual to perceived relative deprivation, this paper assesses the role of visibility in goods and assets vis-à-vis income behind perceptions of relative deprivation. We rely on household survey data that include unique information on reported perceived deprivation with a pre-specified reference group, namely others in the same town or village. Based on a large number of asset and consumption items, we create an index of visible wealth by aggregating visible goods and assets using principal component weights. We show that relative deprivation in visible wealth has a significantly stronger effect than income in determining levels of perceived relative deprivation. The finding is robust under various sensitivity checks and for a number of controls. Our result sheds light on the importance of the visibility of the objects of comparison on an individual’s assessment of the own relative economic situation and proposes that future research should not only rely on income-based deprivation measures.

An Intrusion Detection Method Based on WPSO-SVM and KPCA

With the spread of internet, the applications based on Internet are developing fast. However, it is generally accepted that internet security plays an increasing important role in our information society. As a significant measure, intrusion detection technique seems necessary. The traditional Intrusion Detection System has a high leaking and false alarm rate. We propose a new model based on Kernel Principal Component Analysis and Weight Particle Swarm Optimization-Support Vector Machine in this paper. The Kernel Principal Component Analysis method can be used in extracting the feature of nonlinear datasets, in order to reduce dimension. The Weight Particle Swarm Optimization helps to optimize the parameters in the Support Vector Machine. In the result, the model in this paper has a good performance.

The Role of Visible Wealth for Deprivation

Motivated by the lack of literature linking actual to perceived relative deprivation, this paper assesses the role of visibility in the deprivation of goods and assets vis-a-vis income behind perceptions of relative deprivation. We rely on household survey data that includes unique information on reported perceived deprivation with a pre-specified reference group, namely others in the town or village. Based on a large number of asset and consumption items, we create an index of visible wealth by aggregating visible goods and assets using principal component weights. We find that relative deprivation in visible wealth has a ten percentage point higher explanatory power for reporting a high level of perceived deprivation than that of deprivation in income. The effect is robust under various sensitivity checks and for a number of controls. The finding sheds light on the importance of the visibility of the objects of comparison for an individual's assessment of his relative economic situation and proposes that future research should not solely rely on income-based deprivation measures.

Evaluation and Analysis on Coupling Coordinated Development of Urban Resource, Environment and Economy in Jiangxi Province in China

This paper established a comprehensive appraisement index system of resource and environment and urban economy development and analyzed the coupling coordination of resource, environment and economy development of 11 prefecture-level cities in Jiangxi province in China in 2010 by using the principal component weight method and coupling coordination degree model. The results show that there is a big difference between resource and environment of the 11 cities; economy development is unbalanced; the coupling coordination degrees are overall low between resource, environment and economy development. At last, the authors propose some political advices for the coupling coordinated development of resource, environment and economy.

Apparatus for Implementing 3-Dimensional Virtual Sound and Method Thereof

An apparatus for implementing a 3-dimensional virtual sound and method thereof are disclosed, in which computational and storage complexity are reduced, in which system stability is secured, and by which the 3-dimensional virtual sound can be implemented in such a mobile platform failing to be equipped with expensive instruments for the implementation of the 3-dimensional sound as a mobile communication terminal and the like. The present invention includes a first step of giving an inter-aural time delay (ITD) to at least one input sound signal, a second step of multiplying output signals of the first step by principal component weight, and a third step of filtering result values of the second step by a plurality of low-order approximated IIR filter models of basis vectors extracted from a head related transfer function (HRTF). The basis vectors, extracted from the head related transfer function database are approximated using balanced model approximation technique.

Efficient hardware architecture based on generalized Hebbian algorithm for texture classification

The objective of this paper is to present an efficient hardware architecture for generalized Hebbian algorithm (GHA). In the architecture, the principal component computation and weight vector updating of the GHA are operated in parallel, so that the throughput of the circuit can be significantly enhanced. In addition, the weight vector updating process is separated into a number of stages for lowering area costs and increasing computational speed. To show the effectiveness of the circuit, a texture classification system based on the proposed architecture is designed. It is embedded in a system-on-programmable-chip (SOPC) platform for physical performance measurement. Experimental results show that the proposed architecture is an efficient design for attaining both high speed performance and low area costs.

Gas chromatography/time‐of‐flight mass spectrometry‐based metabonomics of hepatocarcinoma in rats with lung metastasis: elucidation of the metabolic characteristics of hepatocarcinoma at formation and metastasis

Hepatocarcinoma (HCC) has a very high mortality rate and the high recurrence and metastasis rates contribute to the poor prognosis of HCC patients. To understand HCC formation and metastasis, we assessed the metabonomics of rat HCC and HCC with lung metastasis (HLM). The HLM rat model was established by exposure to diethylnitrosamine (DEN). Levels of serum and urine metabolites were quantified with gas chromatography/time-of-flight mass spectrometry (GC/TOFMS), and data were analyzed with partial least-squares discrimination analysis (PLS-DA). Serum and urine levels of some metabolites differed significantly between the control, HCC, and HLM groups. The products and intermediates from glycolysis and glutamate metabolism were elevated, while the tricarboxylic acid (TCA) cycle was inhibited, in both HCC and HLM. HLM samples revealed enhanced metabolism of nucleic acids, amino acids and glucuronic acid. PLS-DA indicated that principal component weighting was greatest for serum serine, phenylalanine, lactic acid, tyrosine and glucuronic acid, and urine glycine, serine, 5-oxyproline, malate, hippuric acid and uric acid. These data provide novel information that will improve understanding of the pathophysiological processes involved in HCC and HLM, and revealed potential metabolic markers for HCC invasion and metastasis.

Head-related transfer function interpolation through multivariate polynomial fitting of principal component weights

Head-related transfer function (HRTF) interpolation plays an important role for implementation of 3D sound system because it can not only reduce the number of measurements for HRTFs, but also reduce the data of HRTFs for seamless binaural synthesis. This paper addresses the problem of accurately realizing the interpolation of HRTF for synthesis of virtual auditory space, and proposes a HRTF interpolation method based on principal component analysis. Firstly, the HRTF is decomposed into principal components and corresponding principal component weights, where principal components are direction-independent and principal component weights are direction- dependent; then the directional variation of the principal component weight is multivariate polynomial fitted with a bivariate function of two spatial angulus (azimuth and elevation). Moreover, a sphere- partitioning optimization scheme is employed to improve the approximation precision. Experiment results demonstrate that HRTFs in the entire sphere surface can be interpolated by the proposed method with small distortion, and the proposed method performs better than conventional methods. Therefore the proposed method gives a promising way for HRTF interpolation.

Dynamic metabonomic analysis of BALB/C mice with different outcomes after D-galactosamine/lipopolysaccharide-induced fulminant hepatic failure

Fulminant hepatic failure (FHF) is one of the most challenging gastrointestinal emergencies encountered in clinical practice. Early identification of patients with FHF who need liver transplantation is very important. To construct a prediction model for the early diagnosis and prognosis of FHF, we studied the dynamics of metabolic intermediates and metabolic profiles with a D-galactosamine (GalN)/lipopolysaccharide (LPS)-treated BALB/c mouse model of FHF. Levels of plasma metabolites were quantified with gas chromatography/time-of-flight mass spectrometry, and data were processed with partial least squares discriminant analysis (PLS-DA). Distinct clustering differences were observed 5 and 6 hours after GalN/LPS treatment between mice that survived and those that died, but there were no differences between these groups 4 hours after treatment. Five hours after treatment, plasma levels of some metabolites differed significantly between the survival, dead, and control groups. Ketogenesis and the tricarboxylic acid cycle were inhibited in both the survival and dead groups, but in the dead group, the urea cycle was also inhibited, and glycolysis was elevated. PLS-DA indicated that principal component weighting was greatest for plasma levels of phosphate, beta-hydroxybutyrate, urea, glucose, and lactate. The Y-predicted scatter plot in the partial least squares (PLS) model assigned samples to the survival or dead groups with an a priori cutoff of 0.10 with 100% sensitivity and specificity. Similar results were observed in 11 FHF patients with different outcomes. In conclusion, the PLS model based on metabonomic analysis can be used to predict outcomes well, and plasma levels of phosphate, beta-hydroxybutyrate, urea, glucose, and lactate may constitute a set of markers for the early diagnosis and prognosis of FHF.

Computational Models of Facial Attractiveness Judgments

We designed two computational models to replicate human facial attractiveness ratings. The primary model used partial least squares (PLS) to identify image factors associated with facial attractiveness from facial images and attractiveness ratings of those images. For comparison we also made a model similar to previous models of facial attractiveness, in that it used manually derived measurements between features as inputs, though we took the additional step of dimensionality reduction via principal component analysis (PCA) and weighting of PCA dimensions via a perceptron. Strikingly, both models produced estimates of facial attractiveness that were indistinguishable from human ratings. Because PLS extracts a small number of image factors from the facial images that covary with attractiveness ratings of the images, it is possible to determine the information used by the model. The image factors that the model discovered correspond to two of the main contemporary hypotheses of averageness judgments: facial attractiveness and sexual dimorphism. In contrast, facial symmetry was not important to the model, and an explicit feature-based measurement of symmetry was not correlated with human judgments of facial attractiveness. This provides novel evidence for the importance of averageness and sexual dimorphism, but not symmetry, in human judgments of facial attractiveness.

Comparison of five allopatric fruit fly parasitoid populations (Psyttaliaspecies) (Hymenoptera: Braconidae) from coffee fields using morphometric and molecular methods

Morphometric studies of five allopatric parasitoid populations (genus Psyttalia Walker) from coffee plantations in Cameroon (Nkolbisson), Ghana (Tafo) and Kenya (Rurima, Ruiru and Shimba Hills) and one non-coffee population (from Muhaka, Kenya) were compared with individuals of Psyttalia concolor (Szépligeti), a species released in several biological control programmes in the Mediterranean Region since the 20th Century. Analyses of wing vein measurements showed the second submarginal cell of the fore wing and its adjoining veins had the heaviest principal component weights and served as the main contributing variables in the diagnostic differentiation of the populations. Two populations (Rurima and Ruiru) were found to be the closest to each other and with the strongest phenetic affinity toward P. concolor (and forming one cluster). Populations from Shimba Hills (of unknown identity), Nkolbisson (P. perproximus (Silvestri)) and Tafo formed a second cluster and were separated from P. concolor. Comparison using amplified fragment length polymorphism (AFLP) also showed the Shimba, Nkolbisson and Tafo populations forming a cluster in a dendrogram generated from their genetic distances, with the Shimba and Tafo populations placed as the most closely related species. Based on consistent morphological similarities, morphometric and ecological data coupled with the genetic evidence from AFLP data, the Shimba population is suggested as belonging to the P. perproximus group and, thus, represents a new occurrence record in Kenya. Our results also support earlier conclusion from cross mating data that populations from Rurima and Ruiru belong to the Psyttalia concolor species-group.

Style‐Based Motion Synthesis†

Abstract Representing motions as linear sums of principal components has become a widely accepted animation technique. While powerful, the simplest version of this approach is not particularly well suited to modeling the specific style of an individual whose motion had not yet been recorded when building the database: it would take an expert to adjust the PCA weights to obtain a motion style that is indistinguishable from his. Consequently, when realism is required, the current practice is to perform a full motion capture session each time a new person must be considered. In this paper, we extend the PCA approach so that this requirement can be drastically reduced: for whole classes of cyclic and noncyclic motions such as walking, running or jumping, it is enough to observe the newcomer moving only once at a particular speed or jumping a particular distance using either an optical motion capture system or a simple pair of synchronized video cameras. This one observation is used to compute a set of principal component weights that best approximates the motion and to extrapolate in real‐time realistic animations of the same person walking or running at different speeds, and jumping a different distance.

Including principal component weights to improve discrimination in data envelopment analysis

This research further develops the combined use of principal component analysis (PCA) and data envelopment analysis (DEA). The aim is to reduce the curse of dimensionality that occurs in DEA when there is an excessive number of inputs and outputs in relation to the number of decision-making units. Three separate PCA–DEA formulations are developed in the paper utilising the results of PCA to develop objective, assurance region type constraints on the DEA weights. The first model applies PCA to grouped data representing similar themes, such as quality or environmental measures. The second model, if needed, applies PCA to all inputs and separately to all outputs, thus further strengthening the discrimination power of DEA. The third formulation searches for a single set of global weights with which to fully rank all observations. In summary, it is clear that the use of principal components can noticeably improve the strength of DEA models.

Fourier Transform Raman Spectra of Linear Low-Density Polyethylene and Prediction of Their Density by Multivariate Data Analysis

Fourier transform Raman spectra have been measured for pellets of sixteen kinds of linear low-density polyethylene (LLDPE) with short branches and one kind of PE without any branch. Before we tried chemometrics analysis, the Raman spectra of LLDPE were investigated by comparing them with the spectrum of PE in order to explore the effects of the branches on the Raman spectra. Partial least-squares (PLS) regression was applied to the Raman spectra in the 1600–600 cm−1 region after multiplicative scatter correction (MSC) to propose a calibration model that predicts the density of LLDPE. The correlation coefficient was calculated to be 0.968, and the root mean square error of cross validation (RMSECV) was found to be 0.0018 g/cm3. The loadings plot of regression coefficients for the calibration model shows that, not only a sharp upward peak at 1417 cm−1 corresponding to the CH2 bending mode reflecting the crystallinity, but also a broad downward peak near 1308 cm−1 corresponding to the amorphous board band of LLDPE plays a key role in the prediction of their density. The chemometrics study has deepened the analysis of the Raman spectra of LLDPE. For example, the detailed analysis of the principal component weight loadings plots has elucidated the existence of bands due to the CH3 groups of branches and those arising from amorphous parts of LLDPE that are almost missing or hidden by other intense bands. In other words, the chemometrics analysis has enhanced spectral resolution.

Non-Destructive Analysis of Photo-Degradation of Poly(Methyl Methacrylate) by near Infrared Light-Fibre Spectroscopy and Chemometrics

Photo-degradation of poly(methyl methacrylate) (PMMA) has been analysed in situ by the use of near infrared (NIR) light-fibre spectroscopy. Fourier transform (FT) NIR spectra have been measured for PMMA plates irradiated by a high-pressure mercury lamp (450 W) for 0, 30, 60, 90, 120, 150, 180 and 210 min. Principal component analysis (PCA) has been carried out for the FT-NIR spectra pretreated by multiplicative scatter correction (MSC) or second derivative. A principal component (PC) weight loadings plot of factor 1 shows that the irradiation time is strongly correlated with the band due to the second overtone of the C=O stretching mode. Partial least squares (PLS) regression has enabled us to propose a calibration model which predicts the irradiation time of the PMMA plates, with a correlation coefficient and the mean square error of cross validation ( RMSECV) of 0.995 and 7.2 min, respectively.

Discrimination of ethylene/vinyl acetate copolymers with different composition and prediction of the content of vinyl acetate in the copolymers and their melting points by near-infrared spectroscopy and chemometrics

Near-infrared (NIR) diffuse reflectance spectra have been measured by use of a rotating drawer for pellets of 12 kinds of ethylene/vinyl acetate (EVA) copolymers with vinyl acetate (VA, the comonomer) varying in the 7–44 wt % range. They are unambiguously discriminated from one another by a score plot of the principal component analysis (PCA) Factor 1 and 2, based upon the NIR spectra pretreated by multiplicative scatter correction (MSC). Principal component (PC) weight loadings for Factor 1 show that the discrimination relies largely upon bands due to the overtone and combination modes arising from the VA unit. We have found one “outlier” in the score plot and elucidated its spectral characteristics based upon PC weight loadings for Factor 2. Partial least-squares (PLS) regression has been applied to propose calibration models which predict the VA content in EVA. The models have been prepared for three kinds of pretreatment, the first derivative, the second derivative, and MSC; and four kinds of wavelength regions. The NIR spectra in the 1100–2200 nm region after the MSC treatment has given the best correlation coefficient and standard error of prediction (SEP) of 0.998 and 0.70%, respectively. The calibration models, prepared by NIR diffuse reflectance spectroscopy for the pellet samples, are compared with previously reported models by NIR transmission spectroscopy for the flowing molten samples, and with those by Raman spectroscopy for the pellet samples. PLS regression has also allowed us to predict melting points of the copolymers with the correlation coefficient and SEP of 0.997 and 0.78°C, respectively. © 1998 John Wiley & Sons, Inc. J Polym Sci B: Polym Phys 36: 1529–1537, 1998

Discrimination of ethylene/vinyl acetate copolymers with different composition and prediction of the content of vinyl acetate in the copolymers and their melting points by near‐infrared spectroscopy and chemometrics

Near-infrared (NIR) diffuse reflectance spectra have been measured by use of a rotating drawer for pellets of 12 kinds of ethylene/vinyl acetate (EVA) copolymers with vinyl acetate (VA, the comonomer) varying in the 7–44 wt % range. They are unambiguously discriminated from one another by a score plot of the principal component analysis (PCA) Factor 1 and 2, based upon the NIR spectra pretreated by multiplicative scatter correction (MSC). Principal component (PC) weight loadings for Factor 1 show that the discrimination relies largely upon bands due to the overtone and combination modes arising from the VA unit. We have found one “outlier” in the score plot and elucidated its spectral characteristics based upon PC weight loadings for Factor 2. Partial least-squares (PLS) regression has been applied to propose calibration models which predict the VA content in EVA. The models have been prepared for three kinds of pretreatment, the first derivative, the second derivative, and MSC; and four kinds of wavelength regions. The NIR spectra in the 1100–2200 nm region after the MSC treatment has given the best correlation coefficient and standard error of prediction (SEP) of 0.998 and 0.70%, respectively. The calibration models, prepared by NIR diffuse reflectance spectroscopy for the pellet samples, are compared with previously reported models by NIR transmission spectroscopy for the flowing molten samples, and with those by Raman spectroscopy for the pellet samples. PLS regression has also allowed us to predict melting points of the copolymers with the correlation coefficient and SEP of 0.997 and 0.78°C, respectively. © 1998 John Wiley & Sons, Inc. J Polym Sci B: Polym Phys 36: 1529–1537, 1998