作者: Axel Plešinger , Bohuslav Růžek , Alena Boušková
关键词:
摘要: We employed multilayer perceptrons (MLP), self organizing feature maps (SOFM), and learning vector quantization (LVQ) to reveal interpret statistically significant features of different categories waveform parameter vectors extracted from three-component WEBNET velocigrams. In this contribution we present discuss in a summarizing manner the results (i) SOFM classification MLP discrimination between microearthquakes explosions on basis single-station spectral amplitude vectors, (ii) SOFM/LVQ recognition initial onset polarities PV'-waveforms, (iii) source mechanism study January 1997 microearthquake swarm based combined multi-station PV-onset polarity SH/PVamplitude ratio (CPA) data.