Paper Info
Title
Experimental comparison of resampling methods in a multi-agent system to assist with property valuation
Abstract
A new version of the a multi-agent system to aid in real estate appraisal, called MAREA-2, was introduced. The system is being developed using Java Spring Framework and is intended for industrial application in cadastral information centres. The major part of the study was devoted to investigate the performance of Bagging, Subagging, and Repeated cross-validation models. The overall result of our investigation was that the majority of models created using resampling techniques provided better or equivalent accuracy than the experts’ method employed in reality. It confirms that automated valuation models can be successfully incorporated into the multi-agent system and be utilized to support appraisers’ work.
Year
DOI
Venue
2011
10.1007/978-3-642-22000-5_36
agent and multi agent systems technologies and applications
Keywords
Field
DocType
major part,overall result,multi-agent system,experimental comparison,java spring framework,new version,automated valuation model,resampling method,repeated cross-validation model,equivalent accuracy,property valuation,industrial application,cadastral information centre,bagging,multi agent systems
Data mining,Real estate appraisal,Cadastre,Computer science,Multi-agent system,Artificial intelligence,Java,Resampling,Valuation (finance),Machine learning
Conference
Volume
ISSN
Citations 
6682
0302-9743
0
PageRank 
References 
Authors
0.34
10
3
Name
Order
Citations
PageRank
Tadeusz Lasota134825.33
Tomasz Łuczak222540.26
Bogdan Trawiński328824.72