The deterministic dendritic cell algorithm with Haskell in earthquake magnitude prediction
- PDF / 743,379 Bytes
- 11 Pages / 595.224 x 790.955 pts Page_size
- 83 Downloads / 149 Views
RESEARCH ARTICLE
The deterministic dendritic cell algorithm with Haskell in earthquake magnitude prediction Wen Zhou1
· Hongbin Dong2 · Yiwen Liang1
Received: 20 March 2019 / Accepted: 8 January 2020 © Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract The earthquake magnitude prediction is a task of utmost difficulty that has been addressed by using many different strategies, with no further transformation thus far. This work evaluates the Haskell based deterministic dendritic cell algorithm (hDCA)’s accuracy when used to predict earthquake magnitude in Sichuan and surroundings. First, eight seismicity indicators have been retrieved from the literature and used as input for the algorithms, and they are calculated from the earthquake catalog of the Sichuan and surroundings by well-known geophysical theory, named Gutenberg-Richter inverse power-law, and characteristic earthquake magnitude distribution and also conclusions drawn by recent related studies. Then, the hDCA is used to predict earthquakes with magnitude larger than 4.5 in the next month. In this work, the proposed method has been compared to the well-known machine learning algorithms, such as Dendritic Cell Algorithm (DCA), Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Back Propagation Neural Network (BPNN), Recurrent Neural Network (RNN), Probabilistic Neural Network (PNN) and Neural Dynamic Classification (NDC). Overall our method yields the promising results in terms of all qualify parameters evaluated. Keywords Earthquake magnitude prediction · Danger theory · hDCA · Geophysical theory
Introduction Earthquake is one of natural disasters that can cause massive casualties and material losses. Human beings cannot stop them, but timely predictions and appropriate security measures can prevent human and material losses. For the non-frequently of earthquake and complexity of the physical process, some scientists believe that earthquakes
Communicated by: H. Babaie Wen Zhou
zw [email protected] Hongbin Dong [email protected] Yiwen Liang [email protected] 1
School of Computer Science, Wuhan University, Wuhan, China
2
School of Cyber Science and Engineering, Wuhan University, Wuhan, China
are unpredictable (Geller et al. 1997), while many others have suggested that it can be predicted (Brehm and Braile 1998; Kirschvink 2000; Allen 1976). Therefore, many seismologists debate whether the earthquake could be predicted on the Nature website in 1999. The consensus between both sides of the debate is actually more than the difference, and both parties agree that at least in terms of existing knowledge, it is impossible to make deterministic predictions of earthquakes reliably and accurately. However, it is possible to calculate the probability of an earthquake that may occur in the future based on the scientific data (Nature 1999). Therefore, several studies have been carried out for earthquake prediction, including the analysis of precursory phenomenon like animal behavior analysis (Grant et al. 2015), soil radon and thoron concen
Data Loading...
