Backward strategy

In the first strategy, a forward strategy, one could start with the feature with the highest A-value and then successively add further features. In the second strategy, a backward strategy, one could start with all features. Then the feature with the lowest discriminating power is deleted first from the entire set, and so on. 

In Tab. 5-13 we report the results of both mentioned strategies for the selection process. In both procedures the WILKS lambda varies monotonously and each set has a significant meaning. We may, therefore, stop the selection process following the misclassification rate. In the forward strategy the first zero error rate appears with the feature set Ti, Mg, Ca in step 3 (Fig. 5-25) whereas in the backward strategy the zero error rate is obtained with the remaining elements Si, Ca, Al, Mg in step 3. Now it is up to the expert to decide which feature set to retain in the future. 

During our previous discussion of strategies for working synthesis problems in Section 8.9, we said that ids usually best to work a problem backward, or retrosyntheticnlly. Look at the target molecule and ask yourself, "What is an immediate precursor of this compound " Choose a likely answer and continue working backward, one step at a time, until you arrive at a simple starting material. Let s try some examples. 

Retrosynthetic (Sections 8.9, 16.11) A strategy for planning organic syntheses by working backward from the final product to the starting material. 

The bottom-up approach contains two distinct stages. First, by successive backpropagation steps one builds a decision policy. Then, this uncovered policy is evaluated and refined, and its expected benefits confirmed before any implementation actually takes place. This two-stage process is conceptually similar to dynamic programming solution strategies, where first a decision policy is constructed by backward induction, and then one finds a realization of the process for the given policy, in order to check its expected performance (Bradley et al., 1977). 

If the nonlinear character of the kinetic law is more pronounced, and/or if more data points than merely the peak are to be used, the following approach, illustrated in Figure 1.18, may be used. The current-time curves are first integrated so as to obtain the surface concentrations of the two reactants. The current and the surface concentrations are then combined to derive the forward and backward rate constants as functions of the electrode potential. Following this strategy, the form of the dependence of the rate constants on the potential need not be known a priori. It is rather an outcome of the cyclic voltammetric experiments and of their treatment. There is therefore no compulsory need, as often believed, to use for this purpose electrochemical techniques in which the electrode potential is independent of time, or nearly independent of time, as in potential step chronoamperometry and impedance measurements. This is another illustration of the equivalence of the various electrochemical techniques, provided that they are used in comparable time windows. 

This is not a problem for guess and check, or working backwards. An organized table could be used, but it would need 25 entries, which would take up time. The strategy will be to start the table, and then look for a pattern ... 

Look for an essay that uses the analysis/classification, comparison and contrast, order of importance, or problem solution strategy. Work backward from the text to create an outline that delineates the organizing structure. 

Based on these devices, different biomass estimation experiments were performed based on the culture fluorescence monitoring and feeding strategy studies were developed as well as bioreactor characterizations via mixing time experiments. During the next years smaller fluorescence probes were developed which could be interfaced with bioreactors via standard electrode ports. These open end detector systems measured the fluorescence fight in the backward di-... 

The only completely rigorous method to determine an optimum subset of variables is to test all possible combinations. The problem with this approach is that even with as few as 100 variables the number of combinations becomes too large to consider. For example, to select an optimum subset of three variables from amongst 100 measurement variables, 161,700 different combinations need to be tested. Because of the large number of possibilities, various strategies have been developed to select variables with a reasonable amount of work. The most common approaches include step forward, step backward,... 

---