Publication: Data analysis technique to extract threshold value for Fuzzy Inference System in classifying malicious agents
Date
2015
Authors
Musa N.A.
Yusoff M.Z.M.
Yusoff Y.
Ismail R.
Journal Title
Journal ISSN
Volume Title
Publisher
Institute of Electrical and Electronics Engineers Inc.
Abstract
This paper describes a data analysis technique used to extract threshold values of a Fuzzy Inference System's (FIS) inputs. In this work, the FIS is used to classify malicious agents' behaviour in a Multi-Agent System (MAS) environment. The extraction of suitable FIS inputs threshold values in the MAS environment are based on data analysis derived from several simulation runs. Subsequently, the extracted threshold values are applied in the rules of the FIS framework for classifying malicious agents' behaviors. Results indicated that extracting the fuzzy threshold values using the data analysis technique is an acceptable alternative method when no domain expert is available. � 2014 IEEE.
Description
Computer forensics; Data handling; Fuzzy systems; Information analysis; Multi agent systems; Data analysis techniques; Domain experts; Fuzzy inference systems; Fuzzy threshold; Malicious agent; Threshold-value; Fuzzy inference