Statistical validation data classification

Fig. 8.11. The classifier development process. Clinical knowledge provides us with a set of classes for supervised classification (top, right). Large numbers of spectra from large sample numbers are reduced to a set of potentially useful features (top, left) or metrics. A modified Bayesian algorithm operates on the metrics to provide predictions that are compared to a gold standard. The end result of the training and validation process is an optimized algorithm, metric set, calibration and validation statistics, and sensitivity analysis of the data...

CM 18.1 to CM 18.3 were assessed in terms of their Cooper statistics, which define an upper limit to predictive performance. In addition, cross-validated Cooper statistics, which provide a more realistic indication of a model s capacity to predict the classifications of independent data, were obtained by applying the threefold cross-validation procedure to the best-sized CTs. In the threefold cross-validation procedure, the data set is randomly divided into three approximately equal parts, the CT is re-parameterized using two thirds of the data, and predicted classifications are made for the remaining third of the data. The cross-validated Cooper statistics are the mean values of the usual Cooper statistics, taken over the three iterations of the cross-validation procedure. The Cooper statistics for CM 18.1 to CM 18.3 are summarized in Table 18.6. 

Chemometrics is a most useful tool in QSAR and QSPR studies, in that it forms a firm base for data analysis and modelling and provides a battery of different methods. Moreover, a relevant aspect of the chemometric philosophy is the attention it pays to the predictive power of the models (estimated by using -> validation techniques), -> model complexity, and the continuous search for suitable parameters to assess the model qualities, such as -> classification parameters and -> regression parameters. Chemometrics includes several fields of mathematics and statistics as listed below. 

The goodness of prediction statistic measures how well a model can be used to estimate future (test) data, that is, how well a regression model (or a classification model) estimates the response variable given a set of values for predictor variables. This statistics is obtained using —> validation techniques. 

In addition to MLP based classifier the presented approach should be validated with other advanced classification techniques such as Support Vector Machine (SVM). The robustness should be further examined with more statistics and with data from real driving scenarios. Inspired by study in [15] driving impairment such as inebriation should be investigated by utilizing bioelectrical impedance with tetrapolar electrode method. Correlation study of bioelectrical impedance and human emotional state can be carried out thereafter. 

In addition to compound quantification, an exhaustive statistical analysis is applied to the data. The classification and/or verification of samples are major objectives. Our approach is, first, to exactly identify the sample by cascading classification models and, second, to validate the sample with respect to its most qualificatory group (e.g., rediluted apple juice concentrate from Poland). This reduces the variance of the validation models and, therefore, increases their discriminatory power. The foundation of the statistical analyses is... 

Several criteria and rules of thumb have been formulated [26,28,46] to answer the question How many PCs In EMDA, criteria based on statistical inference, that is, on formal tests of hypothesis, should be avoided as we do not want to assume, in the model estimation phase, our PCs to follow a specific distribution. In this context, more intuitive criteria, albeit not formal, but simple and working in practice, are preferable, especially graphics-based criteria, such as sequential exploration of scores plots and/or inspection of residuals plots plots of eigenvalues (scree plots [47]) or cumulative variance versus number of components. Different consideration holds when PCA is used to generate data models that are further used, for example, for regression, classification tasks or process monitoring [48,49] (Section 3.1.5), where PCA model validatiOTi, for example, by cross-validation, in terms of performance on the assessment of future samples has to be taken into account. 

---