Parvus: An Extendable Package Of Programs For Data Exploration, Classification And Correlation PARVUS , a software package for general pattern recognition, is specially designed for use on microcomputers. The package consists of a set of fifty programs which offers the experimenter the use of sophisticated statistical techniques in multivariate data analysis, such as Principal Components Analysis. As such, PARVUS can be considered as a comprehensive and efficient laboratory tool that can be readily applied to a large range of problems in data analysis. The modern analytical laboratory provides an abundance of sample information, which necessitates computer-aided reduction and interpretation for efficiency. For example, diode array spectrophotometric methods routinely provide data for many wavelengths per compound, which makes it very difficult. PARVUS can be of great assistance in the evaluation and interpretation of such multivariate data. The programs can be subdivided into several functional groups: data import, data manipulation and preprocessing, feature selection, data processing, classifying methods, correlation analysis, target factor analysis, regression analysis, nonlinear mapping, graphical presentation and utilities. With PARVUS three main groups of problems can be handled (1) explorative analysis and representation, (2) classification and (3) correlation. These are tools for reducing the data set for structuring it, and for finding characteristics in it, as well as for recognizing patterns in the data set. The explorative analysis and representation techniques are useful for the identification of general data features, e.g., to find similarities among the samples or the variables by which they are characterized, to detect anomalies or errors, in order to devise the strategies for classification or correlation. Classification techniques can be applied to all sorts of problems where two or more groups have been defined and the aim is either to evaluate relevance or redundancy of the variables in order to differentiate between the groups, or to assign an unknown sample to one of the known groups. Correlation methods are used to determine the relationship between two variables, or between two groups of variables. Because of its modular structure the techniques implemented in PARVUS can be used in a variety of ways depending on the nature of the data. Every one of the more 50 programs included in the package provides the choice of several options which, when used in combination with other programs, result in numerous different application possibilities, many more than can be appreciated at first glance. The problems that can be tackled with multivariate statistical techniques can be found anywhere. For example, a paleontologist wishes to analyse growth and shape variables in the shell of a species of brachiopods on which he has made a large number of measurements. Or a chemist has analysed several characteristics in the wine from three different districts and he wishes to study the relationships between the characteristics in the hope of being able to determine the origin of an unknown wine sample.

##### Scheda prodotto non validato

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Titolo: | Parvus an Extendable Package of Programs for Data Exploration, Classification and Correlation |

Autori: | |

Data di pubblicazione: | 1988 |

Abstract: | Parvus: An Extendable Package Of Programs For Data Exploration, Classification And Correlation PARVUS , a software package for general pattern recognition, is specially designed for use on microcomputers. The package consists of a set of fifty programs which offers the experimenter the use of sophisticated statistical techniques in multivariate data analysis, such as Principal Components Analysis. As such, PARVUS can be considered as a comprehensive and efficient laboratory tool that can be readily applied to a large range of problems in data analysis. The modern analytical laboratory provides an abundance of sample information, which necessitates computer-aided reduction and interpretation for efficiency. For example, diode array spectrophotometric methods routinely provide data for many wavelengths per compound, which makes it very difficult. PARVUS can be of great assistance in the evaluation and interpretation of such multivariate data. The programs can be subdivided into several functional groups: data import, data manipulation and preprocessing, feature selection, data processing, classifying methods, correlation analysis, target factor analysis, regression analysis, nonlinear mapping, graphical presentation and utilities. With PARVUS three main groups of problems can be handled (1) explorative analysis and representation, (2) classification and (3) correlation. These are tools for reducing the data set for structuring it, and for finding characteristics in it, as well as for recognizing patterns in the data set. The explorative analysis and representation techniques are useful for the identification of general data features, e.g., to find similarities among the samples or the variables by which they are characterized, to detect anomalies or errors, in order to devise the strategies for classification or correlation. Classification techniques can be applied to all sorts of problems where two or more groups have been defined and the aim is either to evaluate relevance or redundancy of the variables in order to differentiate between the groups, or to assign an unknown sample to one of the known groups. Correlation methods are used to determine the relationship between two variables, or between two groups of variables. Because of its modular structure the techniques implemented in PARVUS can be used in a variety of ways depending on the nature of the data. Every one of the more 50 programs included in the package provides the choice of several options which, when used in combination with other programs, result in numerous different application possibilities, many more than can be appreciated at first glance. The problems that can be tackled with multivariate statistical techniques can be found anywhere. For example, a paleontologist wishes to analyse growth and shape variables in the shell of a species of brachiopods on which he has made a large number of measurements. Or a chemist has analysed several characteristics in the wine from three different districts and he wishes to study the relationships between the characteristics in the hope of being able to determine the origin of an unknown wine sample. |

Handle: | http://hdl.handle.net/11581/110109 |

Appare nelle tipologie: | Software o prodotti multimediali |