M.Safi - PH.D. Student, Department of Civil and Environmental Engineering, Amirkabir University of Technology, Tehran, Iran,
M.Tehranizadeh - Professor, Department of Civil and Environmental Engineering, Amirkabir University of Technology, Tehran, Iran,
S.A. Alavinasab - PH.D. Student, Department of Civil Engineering, Tarbiat Modarres University, Tehran, Iran

In this paper an Artificial Neural Network is used for classification of earthquake records of Iran.This classification has been performed using linear acceleration, velocity and displacement response spectra of the records. The spectra have been prepared for a long period range in order to take into account the higher mode effects. Different values of viscous damping have also been examined. The competitive learning neural networks have been employed for this problem,which benefits from the advantages of the unsupervised learning algorithm. Over two hundred records have been used for this problem. The records have been selected from more than 2000,recorded earthquake in Iran. The near field records have been omitted and the accelerograms with medium to high magnitude or minimum four degrees in Richter scale have been selected. The classification results of all three spectra have been performed and compared to each other and the obtained classes have been modified.This has shown the different properties of each type of spectrum. Final classification process has led to the determination of soil type or site characteristics, which have been verified using the existing information for some of the recorded earthquakes. The results of such classifications may be used for the preparation of acceleration, velocity or displacement design spectra