Artificial Intelligence and Neural Networks

Table 6 contains books on artificial intelligence and neural networks as related to chemistry. Also included are books on chemometrics. Table 7 is a compilation of computational chemistry books focused on pertinent areas of physical chemistry crystallography, spectroscopy, and thermodynamics. 

With the advent of powerful computers and easy access to them, and the introduction of expert systems, artificial intelligence, and neural networks in QSAR, radically different models designed from noncongeneric large sets of chemicals have been proposed. No attempts are made to design a model that is easily interpretable in terms of MOA. The main objective of the present models is to provide powerful simulators with a wide domain of application for predicting the toxicity of any kind of molecule. 

The true revolution in analytical robotics will probably come from the development of artificial intelligence. Future robots will be managed by expert systems in such a way that they will be able to find solutions to analytical problems by using both the initial stored information and that obtained from the process it has developed previously. The use of artificial intelligence and neural networks to integrate laboratory automation will be one of the most exciting areas in future robotics research. 

With the enormous amount of data accumulated over a period of more than a century of catalyst development, databases became a natural tool for collecting and analyzing experimental results in catalysis. Based on macroscopic data, such as catalyst composition (heterogeneous catalysts), process variables, and the resultant quality parameters of the product, they allow to inter- and extrapolate the many variables of a real-world catalytic system. More recently, this approach has been expanded to the use of artificial intelligence and neural networks (see Neural Networks in Chemistry) in expert systems. Knowledge-based expert systems, play an important role in the optimization of complex, mostly heterogeneous catalytic systems, where often little is known about the active species of the process. Even when the active species is known and the principle mechanism is well understood on a molecular basis, expert systems provide an important tool for optimizing... 

The classical adaptive control scheme is shown in Figure 2.58. Its goal is to use online identification through artificial intelligence (Al), neural networks, and fuzzy logic to adapt the model to the actual process. Al and model predictive control (MPC) can tolerate inaccuracy and uncertainty in the model, and online training can continuously improve the model. 

Ripley, B. D. (1997b). Statistical ideas for selecting network architectures. In Neural Networks Artificial Intelligence and Industrial Applications (ed. K. B. and S. Gielen), pp. 183-90. Springer. 

Systems such as the one illustrated in figure 23.2 will also incorporate artificial intelligence. The information from the sensors will be used with fuzzy logic and neural networking to enable decisions by individual controllers based on the input from multiple sensors. Such systems will also incorporate sensor self-testing, self-calibration, and fault correction, resulting in reliable, automated systems applicable to any process or production line. These systems will significantly affect the productivity and profitability of food, chemical, and pharmaceutical production. 

Artificial intelligence and simulation have mutually beneficial connections. Model-based expert systems, simulations built out of neural networks, and smart interfaces to simulation tools are commonplace. It is expected that AI and simulation will join forces to model and simulate intelligent behavior (WUdberger 1999). 

James Slagle is a professor in the Department of Computer Science at the University of Minnesota. His research interests are in the areas of artificial neural networks, expert systems, and automated reasoning. He is a fellow of the American Association for the Advancement of Science, American Association for Artificial Intelligence, and the IEEE. 

Choy et al. (2002) use case-based reasoning and neural networks to evaluate and benchmark potential suppliers. Performance of these evaluation methods depends upon data provided by potential suppliers and availability of historical data. The disadvantage of artificial intelligence-based methods is their lack of generality, and subsequently only basic features are usually used. 

There is concern that new technologies such as those often labelled as Artificial Intelligence, and including knowledge-based systems, and neural-networks, will be either excluded from systems, because we do not have the means for their assessment, or be included, but under the counter, beyond even minimal controls. 

Intelligent structures are smart structures that have the added capability of learning and adapting rather than simply responding in a programmed manner, and this is usually accomplished by inclusion of Artificial Neural Network (ANN) into the stmcture (Figure 10.2). 

Several techniques from statistics, such as partial least-squares regression, and from artificial intelligence, such as artificial neural networks have been used to learn empirical input/ output relationships. Two of the most significant disadvantages of these approaches are the following ... 

M. Mulholland, D.B. Hibbert, P.R. Haddad and P. Parslov, A comparison of classification in artificial intelligence, induction versus a self-organising neural networks. Chemom. Intell. Lab. Systems, 30 (1995) 117-128. 

In addition, methods of artificial intelligence (artificial neural networks and genetic algorithms) are applied. 

Another form of artificial intelligence is realized in artificial neural networks (ANN). The principle of ANNs has been presented in Sect. 6.5. Apart from calibration, data analysis and interpretation is one of the most important fields of application of ANNs in analytical chemistry (Tusar et al. [1991] Zupan and Gasteiger [1993]) where two branches claim particular interest ... 

Includes an introduction to artificial intelligence, artificial neural networks, self-organizing maps, and growing cell structures... 

---