Case-based library

As well as purely factual data and production rules, the knowledge base may contain a section devoted to case-based information. This is a library of specific examples that in the past have proved informative and that relate to the area in which the system has expertise. 

For example, the factual component of a knowledge base in an ES whose role is to advise on measures that could be taken to limit the global spread of disease might contain data on the effect of air travel on the movement of disease between countries, rules that describe how the rate of spread of disease depends on the local climate, and so on. The knowledge base might also include case-specific data, such as detailed information about particular instances when a disease has spread unusually quickly from one country to another and the reasons for this 

---

A case-based library for distillation column and distillation sequences synthesis using MINLP has been developed. The retrieval algorithm, including inductive retrieval and nearest neighbor techniques, is presented. The retrieval method and the adaptation of the solution is tested by a heptane-toluene example. 

The aim of case-based reasoning is to provide advice based on a set of known examples that are judged to be relevant to the user s query. Files within the library contain data about past cases relevant to the area of expertise, how they were tackled, what the results of this approach were, and whether the action taken was appropriate and successful. Each case is tagged with a set of attributes that describe the case, so that when the library is searched for relevant material, it can quickly be identified through some form of similarity metric. 

This is the first of two syntheses included in this chapter to highlight how natural product-based scaffolds can be attached to a resin for convenient derivatization into libraries. The other example is the synthesis of sarcodic-tyin libraries described in the next section. In both cases, interesting libraries were made with minimal development time by performing scaffold construction in solution or obtaining it from a natural source. 

The extension of the application of combinatorial chemistry from lead discovery to lead optimization has resulted in a gradual shift from split-pool protocol-based libraries generating mixtures of compounds to the parallel synthesis of discrete analogs (see Fig. 5). The emphasis in the latter case is not on the size of the libraries but rather on the yield, purity,... 

The fitness functions embodied in molecule-based methods are relatively fast, especially when optimisation is based on 2D properties, since they typically involve a pairwise molecular comparison with a target molecule rather than the analysis of an entire library, as is the case in library-based methods. However, there is no guarantee that building libraries from frequently occurring reactants will result in optimised libraries, nor is it possible to optimise properties of the library as a whole. 

Selections and affinity-based methods have also been used to expand the scope of RNA-catalyzed reactions to other classes of reactions including isomerization, carbon-carbon bond forming, metalation, and alkylation reactions. For example, by exploiting the same notions of transition state theory that were used in the initial generation of catalytic antibodies, an RNA was isolated that catalyzes the isomerization of a bridged biphenyl substrate to its diastereomer Fig. 12) [89], In this case a library of 10 random RNA mole-... 

In order to overcome these difficulties, the earlier experiences should be used when solving a new problem. Case-based reasoning (CBR) is an excellent tool for the reuse of experience. In the CBR the most similar case to an actual problem is retrieved from a case library, and the solution of this case is used to solve the actual problem. Finally the solution of the problem is stored in the case library for future use (Aamodt and Plaza, 1994 Watson, 1997). 

The central notion of CBR is a case. The main role of a case is to describe and to remember a single event from past experience where a problem or problem situation was solved. A case is made up of two components problem and solution. Typically, the problem description consists of a set of attributes and their values. Many cases are collected in a set to build a case library (case base). The library of cases must roughly cover the set of problems that may arise in the considered domain of application. 

Due to the large number of available protein structures, kinases still have to be considered the best case for structure-based library design. This is even truer for targeting allosteric modulation [124]. Here, a group of inhibitors has been inden-tified that bind outside the ATP pocket and, thereby, stabilize the DFG motif. 

---