We examine the precursory behavior of geoelectric signals before large earthquakes by means of a previously published algorithm including an alarm-based model and binary classification [H.-J. Chen, C.-C. Chen, Nat. Hazards 84, 877 (2016)]. The original method has been improved by removing a time parameter used for coarse-graining of earthquake occurrences, as well as by extending the single-station method into a joint-stations method. Analyzing the filtered geoelectric data with different frequency bands, we determine the optimal frequency bands of earthquake-related geoelectric signals featuring the highest signal-to-noise ratio. Based on significance tests, we also provide evidence of a relationship between geoelectric signals and seismicity. We suggest using machine learning to extract this underlying relationship, which could be used to quantify probabilistic forecasts of impending earthquakes and to get closer to operational earthquake prediction.

我们通过以前发布的算法（包括基于警报的模型和二进制分类）来检查大地震之前地电信号的前兆行为。Chen C.-C. 陈娜 危害84877（2016）]。通过删除用于地震发生的粗粒度的时间参数，以及通过将单站方法扩展为联合站方法，对原始方法进行了改进。通过分析滤波后的不同频带的地电数据，我们确定了具有最高信噪比的地震相关地电信号的最佳频带。基于显着性检验，我们还提供了地电信号与地震活动性之间关系的证据。我们建议使用机器学习来提取此基础关系，该关系可用于量化即将发生的地震的概率预测并更接近于实际地震的预测。