Development of an Autonomous Distributed MultiAgent Monitoring System for the Automatic Classification of End Users

摘要: The purpose of this study is to investigate the feasibility constructing a software Multi-Agent based monitoring and classification system utilizing it provide an automated accurate for end users developing applications in spreadsheet domain. Resulting in, creation Classification System (MACS). The Microsoft‘s .NET Windows Service agents were utilized develop Monitoring Agents MACS. These function autonomously continuous periodic workbooks by content. Communication Foundation (WCF) Services technology was used together with Oriented Architecture (SOA) approach distribution over World Wide Web order satisfy multiple developer aspect. Prometheus agent oriented design methodology its accompanying Design Tool (PDT) employed specifying designing MACS, Visual Studio.NET 2008 creating agency using visual C# programming language. MACS evaluated against criteria from literature support real-time data collected target group excel developers network. configured execute automatically, without any user intervention as windows service processes web server application system. distributed listen read contents spreadsheets development activities terms file author properties, formulas used, Basic Application (VBA) macro code constructs. Data gathered various resources period time filtered Database Updater Agent residing client This then transfers stores Oracle database via stored procedures further processing that leads developers. Oracle mining algorithms: Naive Bayes, Adaptive Decision Trees, Support Vector Machine analyse results gathering process automate developers. accuracy predictions achieved models compared. comparison showed Bayes classifier best 0.978. Therefore, MACS can be solution high degree accuracy.