Clustering Using Fuzzy Learning Vector Quantization ( LVQ) Methods

This paper is to build models directly from the earthquake data catalog for pattern recognition by numerical computer simulation using Fuzzy LVQ. This paper presents learning vector quantization methods to adapt the size of earthquake clusters to better fit a given data set, especially in the context of non-normalized activations. The basic idea of our approach is to compute a desired radius from the data points that are assigned to a cluster and then to adapt the current radius of the cluster in the direction of this desired radius. Since cluster size adaptation has a considerable impact on the number of clusters needed to cover a data catalog, also examine how to select the number of clusters based on validity measures and, in the context of non-normalized activations, on the coverage of the earthquake data for prediction.This is to recognize the distributed past seismicity on active faults. The evolution of the fault system from the previous earthquakes may prove beneficial to predict the trend of active fault by neural computing, and two examples showed the effectiveness of using a Fuzzy Systems for making identification the indicating precise and accurate forecasting earthquakes epicenter. Our results indicate that (1) Fuzzy LVQ is capable of helping finding certain earthquake clusters in North-West of Iran faults, (2) clustering algorithms could be used for finding a set of potential predictor for classification purposes, and (3) comparison and visualization of the effects of different fault systems is straightforward with the Fuzzy LVQ. In summary, the Fuzzy LVQ provides an excellent format for visualization and analysis of active faults, and is likely to facilitate extraction of useful data information of active fault. The patterns of the observed using SOFM simulation are also comparable with surface geological interpretations.

کلیدواژه ها:

Fuzzy LVQ ، Earthquake prediction ، Clustering ، Earthquake catalog