Nonlinear Canonical Correlation Analysis Research Articles

Non-linear correlation of content and metadata information extracted from biomedical article datasets

Biomedical literature databases constitute valuable repositories of up to date scientific knowledge. The development of efficient machine learning methods in order to facilitate the organization of these databases and the extraction of novel biomedical knowledge is becoming increasingly important. Several of these methods require the representation of the documents as vectors of variables forming large multivariate datasets. Since the amount of information contained in different datasets is voluminous, an open issue is to combine information gained from various sources to a concise new dataset, which will efficiently represent the corpus of documents. This paper investigates the use of the multivariate statistical approach, called Non-Linear Canonical Correlation Analysis (NLCCA), for exploiting the correlation among the variables of different document representations and describing the documents with only one new dataset. Experiments with document datasets represented by text words, Medical Subject Headings (MeSH) and Gene Ontology (GO) terms showed the effectiveness of NLCCA.

Dynamic Modeling of Internet Traffic for Intrusion Detection

Computer network traffic is analyzed via mutual information techniques, implemented using linear and nonlinear canonical correlation analyses, with the specific objective of detecting UDP flooding attacks. NS simulation of HTTP, FTP, and CBR traffic shows that flooding attacks are accompanied by a change of mutual information, either at the link being flooded or at another upstream or downstream link. This observation appears to be topology independent, as the technique is demonstrated on the so-called parking-lot topology, random 50-node topology, and 100-node transit-stub topology. This technique is also employed to detect UDP flooding with low false alarm rate on a backbone link. These results indicate that a change in mutual information provides a useful detection criterion when no other signature of the attack is available.

Deflation based nonlinear canonical correlation analysis

This paper introduces two new techniques for determining nonlinear canonical correlation coefficients between two variable sets. A genetic strategy is incorporated to determine these coefficients. Compared to existing methods for nonlinear canonical correlation analysis (NLCCA), the benefits here are that the nonlinear mapping requires fewer parameters to be determined, consequently a more parsimonious NLCCA model can be established which is therefore simpler to interpret. A further contribution of the paper is the investigation of a variety of nonlinear deflation procedures for determining the subsequent nonlinear canonical coefficients. The benefits of the new approaches presented are demonstrated by application to an example from the literature and to recorded data from an industrial melter process. These studies show the advantages of the new NLCCA techniques presented and suggest that a nonlinear deflation procedure should be considered.

Nonlinear Principal Predictor Analysis: Application to the Lorenz System

Abstract Principal predictor analysis is a multivariate linear technique that fits between regression and canonical correlation analysis in terms of the complexity of its architecture. This study introduces a new neural network approach for performing nonlinear principal predictor analysis (NLPPA). NLPPA is applied to the Lorenz system of equations and is compared with nonlinear canonical correlation analysis (NLCCA) and linear multivariate models. Results suggest that NLPPA is capable of performing better than NLCCA when datasets are corrupted with noise. Also, NLPPA modes may be extracted in less time than NLCCA modes. NLPPA is recommended for prediction problems where a clear set of predictors and a clear set of predictands can be easily defined.

A configuration study on the relationships between production systems and business strategy in the changing environment

This paper describes the development and analysis of a numerical taxonomy of the relationships between production systems and business strategy in the changing competitive environment. The taxonomy was developed with a standard method of cluster analysis, and is based on four types of environment, six types of business strategy and four types of production systems. Eighty-two firms were surveyed through face-to-face interviews. Four kinds of relationships between environment, business strategy, and production systems were observed. The validity of the cluster analysis is supported in the non-linear canonical correlation analysis. Underlying dimensions for the cluster analysis are interpreted as the product or service attribute and the customer focus.

Predictability of Indian Ocean sea surface temperature using canonical correlation analysis

There is strong evidence that Indian Ocean sea surface temperatures (SSTs) influence the climate variability of Southern Asia and Africa; hence, accurate prediction of these SSTs is a high priority. In this study, we use canonical correlation analysis (CCA) to design empirical models to assess the predictability of tropical Indian Ocean SST from sea level pressure (SLP) and SST themselves with lead-times up to one year. One model uses the first twelve empirical orthogonal functions (EOFs) of SLP over the Indian Ocean using different lead-times to predict SST. A CCA model with EOFs of SST as the predictor at the same lead-times is compared to SLP as a predictor and shows the auto-correlation of the system. A CCA using the first five extended empirical orthogonal functions (EEOFs) of sea level pressure over the Indian Ocean basin for an interval of two years combined with SST EOFs as predictors is found to produce the greatest correlation between forecast and observed SSTs. This model obtains higher skill by explicitly considering the development in time of SLP anomalies in the region. The skill of this model, assessed from retroactive forecasts of an 18 year period, shows improvement relative to other empirical forecasts particularly for the central and eastern Indian Ocean and boreal autumn months preceding the Southern Hemisphere summer rainfall season. This is likely due to the limited domain of this model identifying modes of variability that are more pronounced in these areas during this season. Finally, a nonlinear canonical correlation analysis (NLCCA) derived from a neural network is used to analyze the leading nonlinear modes. These nonlinear modes differ from the linear CCA modes with distinct cold and warm SST phases suggesting a nonlinear relationship between SST and SLP over the tropical Indian Ocean.

Nonlinear multivariate and time series analysis by neural network methods

Methods in multivariate statistical analysis are essential for working with large amounts of geophysical data, data from observational arrays, from satellites, or from numerical model output. In classical multivariate statistical analysis, there is a hierarchy of methods, starting with linear regression at the base, followed by principal component analysis (PCA) and finally canonical correlation analysis (CCA). A multivariate time series method, the singular spectrum analysis (SSA), has been a fruitful extension of the PCA technique. The common drawback of these classical methods is that only linear structures can be correctly extracted from the data. Since the late 1980s, neural network methods have become popular for performing nonlinear regression and classification. More recently, neural network methods have been extended to perform nonlinear PCA (NLPCA), nonlinear CCA (NLCCA), and nonlinear SSA (NLSSA). This paper presents a unified view of the NLPCA, NLCCA, and NLSSA techniques and their applications to various data sets of the atmosphere and the ocean (especially for the El Niño‐Southern Oscillation and the stratospheric quasi‐biennial oscillation). These data sets reveal that the linear methods are often too simplistic to describe real‐world systems, with a tendency to scatter a single oscillatory phenomenon into numerous unphysical modes or higher harmonics, which can be largely alleviated in the new nonlinear paradigm.

International Retirement Migration: Retired Europeans Living on the Costa Del Sol, Spain

This paper aims to establish the singularity of the European retired who have moved to live on the Costa del Sol, Spain. To study this topic it was necessary, first, to discuss the constrains that arise when this 'migration' is under close scrutiny; secondly, to depict the geographical model of retired Europeans who move to Spain; and, thirdly, to carry out an ad hoc survey to ascertain the reasons for this residential movement, the advantages and disadvantages of living in Spain after retirement, and the influence on several aspects of the economic and social activity on the receiving area. The survey fieldwork was conducted in April-May 1996. Nonlinear canonical correlation analysis is applied to identify the relationships among demographic and perceptual variables. The results show that nationality and, to a lesser extent, age and academic level are the most important variables for grouping retired Europeans on the Costa del Sol according to their statements. The pleasant Mediterranean climate, the lower cost of living and the informal Spanish lifestyle were the main pull factors of living in Spain, the language acting as the greatest disadvantage. As the retired relate these conditions to life in Spain, a strong relationship with tourist areas is highlighted when the geographical distribution is described.

Nonlinear interdecadal changes of the El Ni�o-Southern Oscillation

Nonlinear interdecadal changes in the El Nino-Southern Oscillation (ENSO) phenomenon are investigated using several tools: a nonlinear canonical correlation analysis (NLCCA) method based on neural networks, a hybrid coupled model, and the delayed oscillator theory. The leading NLCCA mode between the tropical Pacific wind stress (WS) and sea surface temperature (SST) reveals notable interdecadal changes of ENSO behaviour before and after the mid 1970s climate regime shift, with greater nonlinearity found during 1981–99 than during 1961–75. Spatial asymmetry (for both SST and WS anomalies) between warm El Nino and cool La Nina events was significantly enhanced in the later period. During 1981–99, the location of the equatorial easterly anomalies was unchanged from the earlier period, but in the opposite ENSO phase, the westerly anomalies were shifted eastward by up to 25°. According to the delayed oscillator theory, such an eastward shift would lengthen the duration of the warm events by up to 45%, but leave the duration of the cool events unchanged. Supporting evidence was found from a hybrid coupled model built with the Lamont dynamical ocean model coupled to a statistical atmospheric model consisting of either the leading NLCCA or CCA mode.

Kernel-based nonlinear canonical analysis and time reversibility

We consider a kernel-based approach to nonlinear canonical correlation analysis and its implementation for time series. We deduce a test procedure of the reversibility hypothesis. The method is applied to the analysis of stochastic differential equation from high-frequency data on stock returns.

Nonlinear Modes of North American Winter Climate Variability Derived from a General Circulation Model Simulation

Abstract Nonlinear principal component analysis (NLPCA), via a neural network (NN) approach, was applied to an ensemble of six 47-yr simulations conducted by the Canadian Centre for Climate Modelling and Analysis (CCCma) second-generation atmospheric general circulation model (AGCM2). Each simulation was forced with the observed sea surface temperature [from the Global Sea Ice and Sea Surface Temperature dataset (GISST)] from January 1948 to November 1994. The NLPCA modes reveal nonlinear structures in both the winter 500-mb geopotential height (Z500) anomalies and surface air temperature (SAT) anomalies over North America, with asymmetric spatial anomaly patterns during the opposite phases of an NLPCA mode. Only during its negative phase is the first NLPCA mode related to the El Niño–Southern Oscillation (ENSO); the positive phase is related to a weakened jet stream. Spatial patterns of the NLPCA mode for the Z500 anomalies generally agree with those for the SAT anomalies. Nonlinear canonical correlation analysis (NLCCA), also via an NN approach, was then applied to the midlatitude winter GCM data and the observed SST of the tropical Pacific. Nonlinearity was detected in both the forcing field (SST) and the response field (Z500 or SAT) at zero time lag. The leading NLCCA mode for the SST anomalies is a nonlinear ENSO mode, with a 30°–40° eastward shift of the positive SST anomalies during El Niño relative to the negative SST anomalies during La Niña. The leading NLCCA mode for the Z500 anomaly field is a nonlinear Pacific–North American (PNA) teleconnection pattern. The ray path of the Rossby waves induced during El Niño is 10°–15° east of that induced during La Niña. The nonlinear atmospheric response to ENSO is also found in the leading NLCCA mode for the SAT anomalies.

Nonlinear canonical correlation analysis of the tropical Pacific wind stress and sea surface temperature

Nonlinear canonical correlation analysis (NLCCA) formulated by a neural network approach was applied to the monthly surface wind stress (WS) and sea surface temperature (SST) in the tropical Pacific. The strength of the nonlinearity varies with the lead/lag time between WS and SST. Compared to the CCA modes, the NLCCA modes explain more variance of the two sets of variables and have higher canonical correlations, particularly, at longer lead/lag times. Unlike the CCA, the NLCCA modes are capable of capturing the asymmetry in the spatial patterns between El Nino and La Nina episodes in both SST and WS fields. With the WS lagging and then leading the SST, the roles of the predictor field and the lagging response field were interchanged, the spatial asymmetry was found to be considerably stronger in the response field than in the predictor field. Hindcasts for SST (using WS as predictor) show that the NLCCA model is generally slightly better than the CCA model in terms of the correlation skills and the root mean square error, mainly in the eastern equatorial regions (e.g. Nino12 and Nino3). Hindcasts for WS (using SST as predictor) show that despite the relatively rapid decrease of skills with the lead time (compared to the skills of SST prediction from WS), the NLCCA is slightly better than the CCA model, with greater improvement over the western equatorial Pacific.

Hybrid coupled models of the tropical Pacific - II ENSO prediction

Two hybrid coupled models (HCMs), a dynamical ocean model coupled to a nonlinear neural network atmosphere (NHCM), and the same ocean model coupled to a linear regression atmosphere (LHCM), are used to investigate ENSO (El Nino-Southern Oscillation) prediction skills, and to test whether having a nonlinear empirical atmospheric model leads to improvements over HCMs with only a linear atmosphere. Predictions of sea surface temperature anomalies (SSTA) and sea level height anomalies (SLHA) show that (1) although both coupled models can fairly well predict the SSTA and SLHA variations in the equatorial eastern Pacific at lead times of 1 year or so, the LHCM displayed less consistent skills than the NHCM over the decades, with considerably lower skills in the 1990s; (2) both models have better correlation skills in the SLHA predictions than in SSTA predictions for all lead times; and (3) the regions of high correlation skills shifted eastward in the 1990s. A nonlinear canonical correlation analysis of the zonal wind stress anomalies and the SSTA shows that their relation was indeed more nonlinear in the 1990s than in the 1980s, which would give the NHCM an advantage over the LHCM in the 1990s.

The perceived usefulness of different sources of traffic information to trucking operations

Abstract Managers in charge of the California operations of nearly 1200 trucking companies were asked their opinions regarding how useful various sources of traffic information are to their dispatchers and to their drivers. They were also asked to evaluate the usefulness of improved traveler information systems. Nonlinear canonical correlation analysis was used to link company characteristics and perceptions of the value of the sources. Results showed that evaluations of sources such as Internet traffic information, in-vehicle navigation systems, and area-wide dedicated highway advisory radio are primarily related to the location of operations, whether a trucking operation is private or for-hire, the average length of the company's load moves, and the provision of intermodal services.

Nonlinear Canonical Correlation Analysis of the Tropical Pacific Climate Variability Using a Neural Network Approach

Abstract Recent advances in neural network modeling have led to the nonlinear generalization of classical multivariate analysis techniques such as principal component analysis and canonical correlation analysis (CCA). The nonlinear canonical correlation analysis (NLCCA) method is used to study the relationship between the tropical Pacific sea level pressure (SLP) and sea surface temperature (SST) fields. The first mode extracted is a nonlinear El Nino–Southern Oscillation (ENSO) mode, showing the asymmetry between the warm El Nino states and the cool La Nina states. The nonlinearity of the first NLCCA mode is found to increase gradually with time. During 1950–75, the SLP showed no nonlinearity, while the SST revealed weak nonlinearity. During 1976–99, the SLP displayed weak nonlinearity, while the weak nonlinearity in the SST was further enhanced. The second NLCCA mode displays longer timescale fluctuations, again with weak, but noticeable, nonlinearity in the SST but not in the SLP.

Identifying the Overarching Logistics Strategy of Business Processes: An Exploratory Analysis

In a "hyper competitive context", a major issue for enterprises is the strategic use of logistics capability to gain, create and leverage value in the market-place. Management now appreciates that a logistics strategy makes a key contribution to corporate strategy and performance. This paper investigates the extent of complementarity between the logistics strategy and business practices in 20 organisations in the management of their operations, inventory, alliances, customers, supply chain integration, performance assessment, information technology and Electronic Data Interchange (EDI) practices. Non-linear canonical correlation analysis is used to establish the degree of concordance between business practices and the logistic strategy for a sample of Australian organisations involved in freight transportation. The evidence suggests that there is very strong synergy between a logistics strategy and the broader set of business initiatives in transport businesses, giving strong support to the overarching role of logistics strategy. This is in contrast to non-transport organisations where logistics is still seen as a separate activity centre, with the exception of information technology and EDI.

Nonlinear canonical correlation analysis by neural networks

Canonical correlation analysis (CCA) is widely used to extract the correlated patterns between two sets of variables. A nonlinear canonical correlation analysis (NLCCA) method is formulated here using three feedforward neural networks. The first network has a double-barreled architecture, and an unconventional cost function, which maximizes the correlation between the two output neurons (the canonical variates). The remaining two networks map from the canonical variates back to the original two sets of variables. Tested on data sets with correlated nonlinear structures, NLCCA showed that the underlying nonlinear structures could be retrieved accurately under moderately noisy conditions. After a mode had been retrieved, NLCCA was applied to the residual to successfully retrieve the next mode. When tested for prediction skills, the NLCCA outperformed the CCA when the two sets of variables contained correlated nonlinear structures.

Kernel and nonlinear canonical correlation analysis.

We review a neural implementation of the statistical technique of Canonical Correlation Analysis (CCA) and extend it to nonlinear CCA. We then derive the method of kernel-based CCA and compare these two methods on real and artificial data sets before using both on the Blind Separation of Sources.

Changes in ventilatory function in grain processing and animal feed workers in relation to exposure to organic dust.

The effects of organic dust exposure on the configuration of the maximum expiratory flow volume (MEFV) curve was examined multivariately. Data from 390 male workers in the animal feed industry in The Netherlands were analyzed. A multivariate technique called nonlinear canonical correlation analyses was used to study the relationship between a set of organic dust exposure variables and a set of ventilatory function variables. The results indicate an almost independent effect of the overall mean organic dust exposure and the number of years of organic dust exposure on ventilatory function. Increasing mean organic dust exposure was associated with a decrease in both forced expiratory volume in 1 s (FEV1.0) and forced vital capacity (FVC) and decreased flows at high lung volumes only. Increasing number of years of dust exposure was associated with a decrease in FEV1.0 and a decrease in flow at all lung volumes, while the FVC seemed relatively constant. These two distinct patterns of reduction in ventilatory function may represent two different pathological processes. Whereas workers with prolonged exposure showed reduced values for all of the MEFV curve variables, except the FVC, those with only a few years of exposure especially showed a decrease in FVC and peak expiratory flow. The effect of current organic dust exposure was more evident for nonsmokers than for ex-smokers and current smokers. The major finding of this study was an independent effect of overall mean organic dust exposure and the number of years of organic dust exposure on the MEFV curve.

A Reader on Applying Statistics in Public Health and Prevention.

Contributions to this volume discuss the application of statistics in public health and prevention, dealing with subjects in the field of working conditions and occupational health, sexually transmissible disease, dental health, public health tables, the geographical distribution of diseases, intervention, and safe behaviour. Papers dealing with working conditions and occupational health include: F. Andries and C.C.J.H. Bijleveld : Working conditions, strain and stress of automation personnel : analysis of complaints and contentedness; S. van Buuren and J.L.A. van Rijckevorsel : Nonresponse imputation in public health research; A. Bloemhoff : Nonlinear partial canonical correlation analysis of health, social class and work indicators - in comparison with cumulative incidence ratio's; F.H.G. Marcelissen, D.J. van Putten and J.J. Radder : Graphical reference-profiles : a method for analysis and presentation of questionnaire data obtained in Periodic Medical and Workplace Surveys (PMWS).