9658

8902

كارشناسي ارشد

كامپيوتر - معماري كامپيوتر

اصفهان: دانشگاه صنعتي اصفهان، دانشكده برق و كامپيوتر

1393

دوازده،98ص.: مصور،جدول،نمودار

ص.ع.به فارسي و انگليسي

ناصر قديري مدرس

93/12/20

عبدالرضا ميرزايي

مهندسي برق و كامپيوتر

ID8902

Classification of spatial data in order to manage the development of urban regions using MOSES algorithm Amir Hossein Goudarzi Ah goudarzi@ec iut ac ir Date of Submission 2014 9 17 Department of Electrical and Computer Engineering Isfahan University of Technology Isfahan 84156 83111 Iran Degree M Sc Language Persian Supervisor Dr Nasser Ghadiri Modares nghadiri@cc iut ac ir Abstract Spatial data is one of the most important and sensitive elements of social economic andpolitical decision making in life today This is why many needs goals and different organizationalactivities are dependent to the knowledge earned from spatial data which is especially important tostrategic planning Existing Researches in this field are usually neglecting the deep knowledgemined from geographical databases and are based on pure statistical methods So classification ofurban regions may represent a comprehensive basis for land use and finally makes the decisionmaking based on the deep knowledge mined from geographical databases Due to the huge volumeof data gathered in spatial databases mining association rules and high level knowledgerepresentation is challenging For the specific domain of maps and spatial data many spatial datamining algorithms have been proposed However there are few algorithms that can manage bothgeographical and non geographical data using topological means Many decision making problemslike developing urban areas require such perception and reasoning In this thesis an approach basedon genetic programming statistical modelling and knowledge representation is represented Toapply MOSES mining rules considering fuzzy topological relations from spatial data a hybridarchitecture called GGEO benefiting from fuzzy region connection calculus is proposed andimplemented To overcome the problem of time consuming topological relationships calculations this method is based on data preprocessing GGEO analyzes and learns from geographical andnormal data simultaneously and uses topological distance parameters representing a series ofarithmetic spatial formulas as classification rules This approach is resistant to noisy data Howeverall of its stages run in parallel for increasing speed The proposed approach may be used indifferent spatial data classification problems as well as representing an appropriate method of dataanalysis and economic policy making To evaluate the application of the mined knowledge indecision making problems in urban planning domain the method s been used in a highwayplanning problem with limited funds Keywords Spatial data mining Parallel processing of spatial data Fuzzy Regional Connection calculus

ناصر قديري مدرس

عبدالرضا ميرزايي