Rule Improvement Through Decision Boundary Detection Using Sensitivity Analysis
نویسندگان
چکیده
Rule extraction from artiicial neural networks (ANN) provides a mechanism to interpret the knowledge embedded in the numerical weights. Classiication problems with continuous-valued parameters create diiculties in determining boundary conditions for these parameters. This paper presents an approach to locate such boundaries using sensitivity analysis. Inclusion of this decision boundary detection approach in a rule extraction algorithm resulted in signiicant improvements in rule accuracies.
منابع مشابه
A Fuzzy Rule Based System for Fault Diagnosis, Using Oil Analysis Results
Condition Monitoring, Oil Analysis, Wear Behavior, Fuzzy Rule Based System Maintenance , as a support function, plays an important role in manufacturing companies and operational organizations. In this paper, fuzzy rules used to interpret linguistic variables for determination of priorities. Using this approach, such verbal expressions, which cannot be explicitly analyzed or statistic...
متن کاملComparative Analysis of Machine Learning Algorithms with Optimization Purposes
The field of optimization and machine learning are increasingly interplayed and optimization in different problems leads to the use of machine learning approaches. Machine learning algorithms work in reasonable computational time for specific classes of problems and have important role in extracting knowledge from large amount of data. In this paper, a methodology has been employed to opt...
متن کاملAn Iterative Decision Rule to minimize cost of Acceptance Sampling Plan in Machine Replacement Problem
In this paper, we presented an optimal iterative decision rule for minimizing total cost in designing a sampling plan for machine replacement problem using the approach of dynamic programming and Bayesian inferences. Cost of replacing the machine and cost of defectives produced by machine has been considered in model. Concept of control threshold policy has been applied for decision making. If ...
متن کاملTransient Measurement Site Design in pipe networks using the Decision Table Method (DTM)
The accuracy of leak detection and calibration of pipe networks by means of the inverse transient analysis (ITA) is highly affected by the number and location of the measurement sites. This study introduces a conceptual decision-making model, the Decision Table Method (DTM), for the measurement site design of pipe networks with the aim of inverse transient analysis. Through the Decision Table M...
متن کاملUltra-Sensitive Optical Biosensor Based on Whispering Gallery Modes: The Effect of Buffer Solutions Refractive Index on Their Sensitivity and Performance
Background: Whispering gallery modes (WGM) biosensors are ultrasensitive systems that can measure amount of adsorbed layer onto the micro-cavity surface. They have many applications including protein, peptide growth, DNA and bacteria detection, molecular properties measurements and specific interaction and drug table recognitions due to their high sensitivity, compact size and label free sensin...
متن کامل