Research on feature engineering strategy in landslide displacement prediction

Feifei Tang; Tongchuan Wang; Yuzhi Meng; Hailian Zhou; Yun Wan

doi:10.1117/12.2686242

28 July 2023 Research on feature engineering strategy in landslide displacement prediction

Feifei Tang, Tongchuan Wang, Yuzhi Meng, Hailian Zhou, Yun Wan

Proceedings Volume 12756, 3rd International Conference on Applied Mathematics, Modelling, and Intelligent Computing (CAMMIC 2023); 1275613 (2023) https://doi.org/10.1117/12.2686242
Event: 2023 3rd International Conference on Applied Mathematics, Modelling and Intelligent Computing (CAMMIC 2023), 2023, Tangshan, China

Abstract

Due to that redundant feature will degrade the accuracy and efficiency of the displacement prediction model, a feature engineering strategy is proposed in this paper to prompt the displacement prediction. Firstly, the displacement-related factors are sorted out, and these factors are enriched by feature interaction. Then, the decision tree algorithm is combined with Spearman correlation coefficient in feature screening phase to eliminate the redundant features. Finally, based on the feature screening results, an integrated AdaBoost-BP neural network prediction model is constructed. Taking Xinpu landslide in Chongqing as an example, the prediction accuracy of MAE and MSE is 0.234mm and 0.099mm respectively, which performs better than that without feature engineering. It is demonstrated that the proposed feature engineering has superior applicability for landslides prediction.

Citation Download Citation

Feifei Tang, Tongchuan Wang, Yuzhi Meng, Hailian Zhou, and Yun Wan "Research on feature engineering strategy in landslide displacement prediction", Proc. SPIE 12756, 3rd International Conference on Applied Mathematics, Modelling, and Intelligent Computing (CAMMIC 2023), 1275613 (28 July 2023); https://doi.org/10.1117/12.2686242

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available