Hoon-Heng Teh

INSIDE: a connectionist case-based diagnostic expert system that learns incrementally

The authors describe a connectionist case-based diagnostic expert system that can learn while the system is being used. The system, called INSIDE (Inertial Navigation System Interactive Diagnostic Expert), was developed for Singapore Airlines to assist the technicians in diagnosing the inertial navigation system used by the airplanes. The system learns from past repair cases and adapts its knowledge base to newly solved cases without having to relearn all the old cases.<<ETX>>

Imprecise reasoning using neural networks

A logic is defined that weighs all available information and implements it using an emulated neural network. This allows the resulting expert system to be able to learn through examples. It also handles fuzziness in the facts and the rules, as well as the logical operations, in a natural and uniform way. It is more realistic than the certainty factor formalism which leaves out information because it takes the minimum of the certainty factors for and AND operation and maximum of the certainty factors for the OR operation. In the present scheme, all activations are weighted and taken into account. Compared with classical expert systems, the present system has the advantage of operating in two modes. In the normal mode, rules are given by experts and weights are assigned values. In the learning mode, weights are allowed to vary while the system is fed with examples.<<ETX>>

Utilizing a neural logic expert system in currency option trading

Abstract International currency market weathers all forms of worldwide current affairs and governmental economics statistics releases. Such news causes fluctuations in demands and supplies for major currencies. In order to gain from currency tradings, a trader has to gauge, sometimes guess using gut feelings, the likely market movement, and act accordingly. In this paper, a neural network-based expert decision support system for assisting users in making currency option trading decisions is presented. By utilizing neural network technology in its inference engine, the system is able to learn new knowledge and refine its inference strategy through usage. Furthermore, it can change its reasoning strategies according to different users, thus realizing the idea of personalized logic system.

A graphical operating environment for neural network expert systems

A window-based platform, known as the Graphical Environment for Neuronet Expert Systems (GENES), is proposed. The platform provides the user with an easy-to-learn, easy-to-use operating environment for creating, training, editing, and enhancing neural-network-based expert systems. The underlying neural logic network (NELONET) has been shown to be capable of doing logical inferencing and is used in two large-scale-operation expert systems. Building on top of the X-window system and the OPENLOOK user interface, GENES inherits the select-and-perform operation strategy for neural network objects. The systems knowledge base contains simple network elements that correspond to rules in a conventional system. During the inference process, these network elements are linked up dynamically to form a large neural network which will operate according to the NELONET activation rules.<<ETX>>

INSIDE: a neuronet based hardware fault diagnostic system

An inertial navigation system interactive diagnostic expert (INSIDE) was developed for troubleshooting an avionic line-replaceable unit, the inertial navigation system. INSIDE was designed based on a neural network model called neural-logic network. The knowledge base can be constructed using a neural-logic network by learning from past cases recorded in the workshop log book. To complement the connectionist knowledge base, a flowchart module which captures the knowledge of troubleshooting flowcharts was also implemented as part of the system. During operation, if the connectionist module fails to derive the solution, the user will be directed to the flowchart module for guidance. After the case is solved, it can be captured as a new example to be acquired by the connectionist module. Besides providing an economical way for developing fault diagnostic systems in general, the learning process of the system highly resembles the way an expert acquires knowledge through experience

Heuristic solutions for the general maximum independent set problem with applications to expert system design

Proposes a set of heuristic algorithms for the solutions of the general maximum independent set problem. These include the commonsense heuristic algorithm, the first-ratio heuristic algorithm, the higher-order ratio heuristic algorithm, the peak heuristic algorithm, and the grouping heuristic algorithm. It is also shown how these algorithms may be used to improve the efficiency of expert system design.<<ETX>>

Rule-based reasoning using extended neural logic network

Neural logic network (NEULONET) are studied in National University of Singapore to incorporate both the pattern processing capability of multilayer perceptrons and the logical inference capability of Boolean logic inference networks within a single frame of neural network environment. In this paper, a few extensions to the NEULONET are proposed. These enhancements to the network structure strengthen its ability to perform rule-based reasonings. The concept of network element (netel) is introduced. With netels, expert system rules may now be easily mapped into rudimentary NEULONETs. The resulting netel knowledge base inherits the semantic meanings of the expert system rules and the learning ability of the connectionist architecture.

Multi-valued neural logic networks

Two types of networks that are useful in developing expert systems are proposed. The probabilistic network can be used for predictive types of expert systems, whereas the fuzzy network is more suitable for expert systems that help in decision-making. In both cases, the expert system can operate in two modes. In the normal mode, rules are given by experts and weights are assigned values. In the learning mode, weights are allowed to vary while the system is fed with examples.<<ETX>>

Fuzzy neural-logic system

A realization of fuzzy logic by a neural network is described. Each node in the network represents a premise or a conclusion. Let x be a member of the universal set, and let A be a node in the network. The value of activation of node A is taken to be the value of the membership function at point x, m/sub A/(x). A logical operation is defined by a set of weights which are independent of x. Given any value of x, a preprocessor will determine the values of the membership function for all the premises that correspond to the input nodes. These are treated as input to the network. A propagation algorithm is used to emulate the inference process. When the network stabilizes, the value of activation at an output node represents the value of the membership function that indicates the degree to which the given conclusion is true. Weight assignment for the standard logical operations is discussed. It is also shown that the scheme makes it possible to define more general logical operations.<<ETX>>

Neural logic hybrid system for currency option trading

International currency market is very sensitive to all forms of worldwide current affairs and governmental economics statistics releases. Such news frequently causes imbalances in the demand and supply of currencies and therefore leads to fluctuations of the exchange rates. In the past, foreign exchange traders depend heavily on their experience and gut feelings when they transact. The emerging neural-network based expert systems has become a viable tool for supporting such strategic decision-making. Besides being flexible, such as having learning capability, these systems can also handle the fuzziness and biasness of human decision processes. This paper presents a neural network expert system for supporting the forex trader in their decision processes.