A Protocol and Tool for Developing a Descriptive Behavioral Model

ECU Author/Contributor (non-ECU co-authors, if there are any, appear on document)

Steven Jeck (Creator)

Institution

East Carolina University (ECU )

Web Site: http://www.ecu.edu/lib/

Advisor

John Placer

Abstract: Fuzzy rules have been used to model complex human behavior in order to develop sophisticated industrial control systems. The use of fuzzy rules to create a behavioral model provides a quantitative basis for discussing the contribution of elements of the model to theories about the behavior. The application of a protocol and tool simplifies the development of a behavioral model from observational data. Extraction of a high level linguistic behavioral model from the observational data is used to discover knowledge about the data. Tuning of the model is accomplished by parameter optimization through the adjustment of membership functions using the genetic fuzzy self-adaptive system. A case study demonstrating the use of the protocol and tool is presented. In the study a behavioral model is developed that integrates the analysis of the observational data with Social Network Analysis. The integrated behavioral model provides an effective platform for a quantitative analysis of the activities impacting behavior. 

Additional Information

Publication

Thesis

Date: 2012

Keywords

Artificial intelligence, Computer science, Computer engineering, Fuzzy Rule Based System, Genetic Algorithm, Genetic Fuzzy Systems, Linguistic Model, Social Network Analysis

Subjects

Human behavior models

Data mining

Email this document to