Propose Design Of Merapi Volcano Eruption Prediction Using Hybrid Time Series And Machine Learning

Volcanic eruptions exemplify natural disasters. Consequently, surveillance is crucial for identifying volcanic activity and collecting information on seismic events during eruption and non-eruption periods. Volcano monitoring involves the identification of seismic occurrences. Accurate identification of seismic events necessitates dependable and efficient seismic event detection and classification. Automated event detection offers benefits and time savings in seismic data processing compared to manual interpretation before event categorization. Seismic data waveforms obtained from observation stations are utilized to detect events. This work employs a time series algorithm and novel hybrid approaches to forecast volcanic eruptions. Hybrid time series classify true or false data from seismic events. At the same time, machine learning algorithms are employed to classify labelling signals based on the particular type of seismic events. The final proposed model can be used to predict volcano eruption precisely.