Abstract: The invention relates to a method for the identification of an analyte in a sample fluid. The method comprises the performance of a set of measurements on at least two sensors of different types. Sensors of different types may comprise Gas Chromatography sensors, CMOS gas sensors, a combination thereof, or more generally sensors in which it is possible to control a first physical parameter and measure, for different values of the first parameter, a second physical parameter representative of at least a concentration of the analyte. The method comprises the detection of maxima of the second parameter, the access to databases in order to find reference values of the first parameters in which maxima are found, and the comparison of values at which maxima are found and reference values, on the two sensors.

Abstract: A method and apparatus are provided for characterizing a product sample for example in comparison to a reference sample using a sensor such as a gas chromatograph or a MOS sensor. This characterization may comprise an indication of whether or not the product sample conforms to a quality criterion. The comparison of the sensor output measurements for the product sample is compared to maximum and minimum value curves, which may be derived from measurements of the reference sample, whereby adjacent samples outside the envelope defined by these maximum and minimum values are grouped together. A dissimilarity index may be determined for the anomalous values as a whole, or on a per group basis. The groups may be classified depending on the shape they describe, in particular the presence, or not, of peaks, and correspondingly the shape of the corresponding part of the envelope.

Abstract: The invention relates to a method for the identification of an analyte in a sample fluid. The method comprises the performance of a set of measurements on at least two sensors of different types. Sensors of different types may comprise Gas Chromatography sensors, CMOS gas sensors, a combination thereof, or more generally sensors in which it is possible to control a first physical parameter and measure, for different values of the first parameter, a second physical parameter representative of at least a concentration of the analyte. The method comprises the detection of maxima of the second parameter, the access to databases in order to find reference values of the first parameters in which maxima are found, and the comparison of values at which maxima are found and reference values, on the two sensors.

Abstract: A method and apparatus are provided for characterizing a product sample for example in comparison to a reference sample using a sensor such as a gas chromatograph or a MOS sensor. This characterization may comprise an indication of whether or not the product sample conforms to a quality criterion. The comparison of the sensor output measurements for the product sample is compared to maximum and minimum value curves, which may be derived from measurements of the reference sample, whereby adjacent samples outside the envelope defined by these maximum and minimum values are grouped together. A dissimilarity index may be determined for the anomalous values as a whole, or on a per group basis. The groups may be classified depending on the shape they describe, in particular the presence, or not, of peaks, and correspondingly the shape of the corresponding part of the envelope.

Abstract: The comparative analysis of a sample, derived from a product, with respect to a database comprises the step of determining the class membership of the different characteristics (variables) describing the samples: whether they are characteristics common to the sample under test and the database, they are characteristics particular to the sample under test, or they are characteristics particular to the database. The assignment of the variables to these classes enables parameters to be defined for global comparison of the sample under test and the database, based on: ratios summarizing the values taken by the variables of the different classes, or the distribution of the variables of the different classes. A machine implementation of the analysis is described.

Abstract: The comparative analysis of a sample, derived from a product, with respect to a database comprises the step of determining the class membership of the different characteristics (variables) describing the samples: are they characteristics common to the sample under test and the database, are they characteristics particular to the sample under test, or are they characteristics particular to the database? The assignment of the variables to these classes enables parameters to be defined for global comparison of the sample under test and the database, based on: ratios summarizing the values taken by the variables of the different classes, or the distribution of the variables of the different classes.

Abstract: The intensity of an odour can be quantified by determining what is the response of an odour sensing device to that odour, then transforming the response data to an odour intensity value based on transformation data relating to a set of selected reference odours. The transformation data includes organoleptic data indicating how odour intensity values assigned to the set of reference odours by a sensory panel depend upon the concentration of the reference compounds and includes data indicating how the response data of the odour sensing device when exposed to the set of reference odours depends upon the concentration of the reference compounds. The reference odours may be basic odours defining the dimensions of a multi-dimensional space in which odours can be defined.

Abstract: The intensity of an odour can be quantified by determining what is the response of an odour sensing device to that odour, then transforming the response data to an odour intensity value based on transformation data relating to a set of selected reference odours. The transformation data includes organoleptic data indicating how odour intensity values assigned to the set of reference odours by a sensory panel depend upon the concentration of the reference compounds and includes data indicating how the response data of the odour sensing device when exposed to the set of reference odours depends upon the concentration of the reference compounds. The reference odours may be basic odours defining the dimensions of a multi-dimensional space in which odours can be defined.

Abstract: The invention concerns a classifying apparatus, used in particular for recognising or characterising odours, applying a combination of statistical methods and neuronal networks to classify into the appropriate class instances of a plurality of different classes presented to the apparatus. The apparatus comprises a processing unit for determining for each class (j) a subspace (SEj) wherein the instances of said class are best separated from instances of other classes, said subspace being defined by synthetic variables (VDj), and for determining a discriminating subspace (SED) defined by the whole set of synthetic variables identified for the whole set of classes. Each neuron in the neuronal input layer corresponds to one of the variables defining the discriminating space (SED) and each neuron of the output layer corresponds to one of the classes.

Abstract: A classification apparatus, notably for uses in recognition or characterisation of odors, comprises a plurality of sensors for generating raw data representing a plurality of instances of a plurality of different classes; and a processing unit for processing said raw data so as to determine an identification model. The identification model comprises definitions of the classes and a particular allocation rule selected dependent upon the application. The class definitions are established by analysing data obtained during a learning phase, the analysis being performed according to a particular information extraction method. During a later identification phase, the identification model enables an instance of unknown class to be allocated to an appropriate class amongst those defined during the learning phase. The information extraction method can be selected dependent upon the application.