William Cheetham

Case-Based Reasoning in Color Matching

A case-based reasoning system for determining what colorants to use for producing a specific color of plastic was created. The selection of colorants needs to take many factors into consideration. A technique that involved fuzzy logic was used to compare the quality of the color match for each factor. The system has been in use for two years at a growing number of GE Plastics sites and has shown significant cost savings.

Measures of Solution Accuracy in Case-Based Reasoning Systems

The case-based reasoning (CBR) methodology can be augmented with the ability to determine the confidence in the correctness of individual solutions. A confidence calculation can be added to the REUSE portion of the CBR methodology. The confidence calculation takes confidence indicators, like “number of cases retrieved with best solution” and “average similarity of cases which suggest an alternative solution,” and generates a confidence value. The information gain algorithm C4.5 can be used to select the best confidence indicators by evaluating their usefulness in historical cases. A genetic algorithm can be used to optimize and maintain the confidence calculation.

Case-Based Reasoning with Confidence

A case-based reasoning system can produce both a solution and an estimate of the confidence in that solution. The confidence value can be used to determine whether the solution does or does not have the needed accuracy. A statistical method can be used to compute a confidence value from information generated during the case-based reasoning process. This confidence value allows users to know when the results of the system should and should not be used.

SOFT-CBR: a self-optimizing fuzzy tool for case-based reasoning

A generic Case-Based Reasoning tool has been designed, implemented, and successfully used in two distinct applications. SOFT-CBR can be applied to a wide range of decision problems, independent of the underlying input case data and output decision space. The tool supplements the traditional case base paradigm by incorporating Fuzzy Logic concepts in a flexible, extensible component-based architecture. An Evolutionary Algorithm has also been incorporated into SOFT-CBR to facilitate the optimization and maintenance of the system. SOFT-CBR relies on simple XML files for configuration, enabling its widespread use beyond the software development community. SOFT-CBR has been used in an automated insurance underwriting system and a gas turbine diagnosis system.

A Phylogenetic Analysis of Aerobic Polychlorinated Biphenyl-Degrading Bacteria

Several bacterial isolates were characterized based on their abilities to degrade specific polychlorinated biphenyls (PCBs) and their 16S rRNA gene sequences. The members of one group of bacteria consisting of Alcaligenes species, including the PCB-degrading bacterium Alcaligenes eutrophus H850, had strong abilities to degrade a broad range of PCBs but not the di-para-chlorine-substituted congeners. The members of another group, which included the PCB-degrading bacterium originally classified as Corynebacterium sp. strain MB1, had strong abilities to degrade di-para-chlorine-substituted PCBs. These bacteria were most likely different members of Rhodococcus species.

Fuzzy case-based reasoning for decision making

A methodology is described for an automated decision making process. The output of the decision process is a solution to the decision and a confidence value in the correctness of the solution. A fuzzy case-based reasoning approach is used to determine the solution. A fuzzy confidence engine uses values that are by-products of determining the solution in order to calculate the confidence value. This methodology has been implemented in a number of applications.

Fielded applications of case-based reasoning

This commentary describes notable commercial applications of case-based reasoning, including systems that have been in continuous profitable use for over a decade. It is divided into sections on engineering applications, helpdesk applications and on-line case-based reasoning.

Automating the underwriting of insurance applications

An end-to-end system was created at Genworth Financial to automate the underwriting of Long Term Care (LTC) and Life Insurance applications. Relying heavily on Artiticial Intelligence techniques, the system has been in production since December 2002 and today completely automates the underwriting of 19.2% of the LTC applications. A fuzzy logic rules engine encodes the underwriter guidelines and an evolutionary algorithm optimizes the engines performance. Finally, a natural language parser is used to improve the coverage of the underwriting system.

Soft case-based reasoning

The aim of this commentary is to discuss the contribution of soft computing—a consortium of fuzzy logic, neural network theory, evolutionary computing, and probabilistic reasoning—to the development of case-based reasoning (CBR) systems. We will describe how soft computing has been used in case representation, retrieval, adaptation, reuse, and case-base maintenance, and then present a brief summary of six CBR applications that use soft computing techniques.

ELSI: A Medical Equipment Diagnostic System

A case-based reasoning system for diagnosing medical equipment, called ELSI, has been in use by the GE corporation since 1994. When a customer or field engineer calls the service center for help with a problem, the equipments error log is automatically downloaded. In ninety seconds or less, ELSI displays a sorted list of the best-matching logs in a case base of previous known problems, shows the fix, service notes, explains which sections of the log match, and which fixes each section predicts. This diagnostic information allows the service center engineer to recommend a temporary work-around or remote fixes to a customer, or helps a field engineer show up on site with the right parts the first time.