基于机器学习的地震异常数据挖掘模型

The Seismic Anomaly Data Mining Model Based on Machine Learning

研究基于机器学习的地震异常数据挖掘方法。在进行地震异常数据挖掘过程中,由于地震监测系统信号时变性及监测环境的不稳定性,采用传统的方法进行挖掘,其挖掘的精确度较低。为此,提出基于机器学习的地震异常数据挖掘方法。根据机器学习的相关理论获取标准方程组和最小均方误差值,实现异常数据挖掘最优模型的构建,通过计算数据的特征向量,建立地震监测数据特征库,依据获取的概率值实现对监测数据的正确判断,从而完成对地震异常数据的有效挖掘。实验结果表明,利用基于机器学习的地震异常数据挖掘方法,能够有效的提高地震异常数据的挖掘准确度与挖掘效率,保证了地震监测系统的有效性。

Research on seismic anomaly data mining method based on machine learning. In seismic anomaly data mining process, because of the earthquake monitoring system signal time-varying instability and monitoring environment, the traditional approach to mining, the mining accuracy is low. For this, put forward an abnormal data mining method based on machine learning. Based on the related theory of machine learning for standard equations and minimum mean square error values, to achieve optimum abnormal data mining model building, through calculating the data of characteristic vector and the earthquake monitoring data in feature library, based on the monitoring data of probability value to obtain the correct judgment, thus accomplishes the effective mining of earthquake anomaly data. The experimental results show that the use of seismic anomaly data mining method based on machine learning, and effectively improved the accuracy of seismic anomaly data mining and the mining efficiency, ensure the effectiveness of the earthquake monitoring system.