Predicting Levels Of Damages To Buildings Caused By Earthquake

• Author(s):

  Arulselvam C | Gowtham R | Pranavvarshan AT | Rajadurai P | Dr.R.Ragupathy

• Keywords:

  Earthquake, Building Damage, Deep Learning, CNN, BLSTM, GBNN, TabNet, TabPFN, NODE, Nepal Earthquake Dataset

• Abstract:

  Earthquakes Are Among The Most Devastating Natural Disasters, Causing Extensive Damage To Infrastructure And Posing Severe Threats To Human Life. Predicting The Level Of Building Damage Resulting From An Earthquake Can Greatly Enhance Disaster Preparedness And Response. This Project Explores The Use Of Advanced Deep Learning Techniques—such As CNN, BLSTM, GBNN, TabNet, TabPFN, And NODE—for Classifying Damage Levels Into Three Categories: Low, Medium, And High. Using The 2015 Nepal Earthquake Dataset, Which Includes Over 25,000 Records And 39 Features, Our Models Demonstrate Improved Performance, Achieving Accuracy Rates Of Over 74.5% In Some Cases. These Findings Highlight The Potential Of Deep Learning For Effective Structural Damage Assessment.

---

Other Details

• Paper id:

  IJSARTV11I4103219

• Published in:

  Volume: 11 Issue: 4 April 2025

• Publication Date:

  2025-04-20

Download Article