首页    期刊浏览 2024年11月24日 星期日
登录注册

文章基本信息

  • 标题:ALTERNATIVE APPROACH TO SEISMIC HAZARDS PREDICTION USING NON PARAMETRIC ADAPTIVE REGRESSION METHOD
  • 本地全文:下载
  • 作者:DADANG PRIYANTO ; MUHAMMAD ZARLIS ; HERMAN MAWENGKANG
  • 期刊名称:Journal of Theoretical and Applied Information Technology
  • 印刷版ISSN:1992-8645
  • 电子版ISSN:1817-3195
  • 出版年度:2020
  • 卷号:98
  • 期号:21
  • 页码:3425-3435
  • 出版社:Journal of Theoretical and Applied
  • 摘要:Research with data mining processes to find certain patterns related to mathematical functions such as Correlation, Classification and regression associations, Clustering and others are grouped into two categories namely Descriptive data mining and Predicted data mining. Data mining process Prediction to find out the relationship between variables can be used Parametric and Non-Parametric methods. Many non-parametric methods used one of them is the Multivariate Adaptive Regression Spline (MARS) method. The flexible nature of MARS modeling can be applied to various fields of application including earthquake prediction research. Research on earthquakes contains many parameters that are definitely necessary to get optimal results with cone optimization models difficult to do this research was conducted to complete research on earthquake predictions with uncertain parameters. This study uses a non-parametric method with MARS and to improve its ability to use the CMARS model which is the back of the MARS algorithm. The results of this study after observing the testing of parameters with a combination of basis functions (BF), Maximum Interaction (MI) and Minimum Observation (MO) obtained the results of predictive analysis with a mathematical model that has two basis functions (BF) namely MODEL (PGA) = BF1, BF2, BF3, BF5, BF7, BF9, BF10, BF11, BF13, BF14, BF15, and BF16. The model was obtained from trial and error observations with a combination of basis functions (BF) = 16, MI = 2, and MO = 2. Based on the level of importance of the independent variables on the dependent variable is the Epicenter Distance (R-epi), Magnitude (Mw), Temperature of the incident location (SUHU), and Depth (Depth). The results of the prediction analysis can reveal six areas that have the highest level of earthquake hazard in Lombok, namely the first area of Malacca, North Lombok Regency (KLU), first Genggelang, Ganga (KLU), Tegal Maja, Tanjung, Winner (KLU), Senggigi choice, Malimbu Regency, West Lombok (Lobar), Mataram, as many as Senggigi, Malimbu (Lobar), and the sixth are Mangsit, and Senggigi (Lobar).
  • 关键词:Non Parametric;Prediction Analysis;MARS;C-MARS;PGA;Data Mining
国家哲学社会科学文献中心版权所有