Feature selection Subject

In Section 33.2.2 we showed how LDA classification can be described as a regression problem with class variables. As a regression model, LDA is subject to the problems described in Chapter 10. For instance, the number of variables should not exceed the number of objects. One solution is to apply feature selection or... 

In this chapter we have been able to discuss only some of the more common and basic methods of feature selection and extraction. This area is a major subject of active research in chemometrics. The effectiveness of subsequent data processing and interpretation is largely governed by how well our analytical data have been summarized by these methods. The interested reader is encouraged to study the many specialist texts and journals available to appreciate the wide breadth of study associated with this subject. 

Animal models of disease that might be considered for use in SP shonld be dictated by whether or not they improve predictive value of adverse reactions in patients, bnt of course they may also be valuable/preferred in studies on mechanism of action and efficacy. Scientists may be guided by selection of specific models that are well known to manifest increased sensitivity for adverse reactions in the clinic (e.g., obesity, senility, heart failure, hypertrophy, diabetes). Clinicians already know, and consider in design of clinical trials, many features of subjects (e.g., age, geographical location, sex, somatotype, strain) that inflnence the outcome of smdies. It is likely that diversity of subjects in preclinical smdies is as important as diversity of snbjects in clinical studies. My suspicion is that smdies conducted on one mouse, one rat, one rabbit, one gninea pig, one dog, and one monkey per each of four groups (vehicle, low dose, mid... 

Chapters 9 and 10 deal with preprocessing of original data and feature selection. These problems must be treated at the beginning of a pattern recognition application. These Chapters have not been positioned at the beginning of the text because a more detailed description of these subjects requires some basic knowledge of pattern recognition methods. 

There are four basic predictive human sensitization tests in current use (1) a single-induction/single-challenge patch test (2) repeated-insult patch test (RIPT) (3) RIPT with continuous exposure (modified Draize) and (4) the maximization test all of these use similar customized patches (Frosch and Kligman 1979 Kaminsky et al. 1986). Principal features of human sensitization assays are summarized in Table 1, and further details can be found in MarzuUi and Maibach (1996). For assays other than maximization, 150-200 subjects are usually tested. Henderson and Riley (1945) statistically showed that if no positive reactions are observed in 200 randomly selected subjects, as many as 15/1000 of the general population may react (95% confidence). As sample size is reduced, the likelihood of unpredicted adverse reactions in the general population increases. 

Sets of spectroscopic data (IR, MS, NMR, UV-Vis) or other data are often subjected to one of the multivariate methods discussed in this book. One of the issues in this type of calculations is the reduction of the number variables by selecting a set of variables to be included in the data analysis. The opinion is gaining support that a selection of variables prior to the data analysis improves the results. For instance, variables which are little or not correlated to the property to be modeled are disregarded. Another approach is to compress all variables in a few features, e.g. by a principal components analysis (see Section 31.1). This is called... 

This article is an attempt at evaluating new important features of tin(II) chemistry the central point is the interrelationship between molecular structure and reactivity of molecular tin(II) compounds. To define these compounds more closely, only those are discussed which are stable, monomeric in solvents and which may be classified as carbene analogs21. Thus, not a complete survey of tin(II) chemistry is given but stress is laid on the structures and reactions of selected compounds. A general introduction to the subject precedes the main chapters. For comparison, also solid-state tin(II) chemistry is included to demonstrate the great resemblance with molecular tin(II) chemistry. Tin(II) compounds, which are either generated as intermediates or only under definite conditions such as temperature or pressure, are not described in detail. 

---