The use of AI is becoming increasingly important for predicting natural disasters like landslides (Representational image)
The developed algorithms have been tested for landslides and can be applied to other natural phenomena such as floods, avalanches, extreme weather events, rock glaciers and permafrost.
Indian Institute technology Researchers at Mandi have developed a new algorithm using Artificial Intelligence and Machine Learning (AI & ML) that can improve the accuracy of natural hazard predictions. Algorithm developed by Dr. Deriks Praj Shukla, Associate Professor, School of Civil and Environmental Engineering, IIT Mandi and Dr. Sharad Kumar Gupta, former Research Scholar, IIT Mandi, currently working at Tel Aviv University (Israel) can deal with these problems. The challenge of data imbalance for landslide susceptibility mapping that represents the probability of landslide events in a given area. The results of their work have recently been published in the journal Landslide.
A landslide susceptibility mapping (LSM) indicates the likelihood of landslides occurring in a specific area based on causative factors such as slope, elevation, geology, soil type, distance from faults, rivers and faults, and historical landslide data.
The use of AI is becoming increasingly important for predicting natural disasters such as landslides. They can potentially predict extreme events, create hazard maps, detect events in real time, provide situational awareness and aid decision making. ML is a subfield of AI that enables computers to learn and improve from experience without being explicitly programmed. It is based on algorithms that can analyze data like human intelligence, identify patterns and make predictions or decisions.
ML algorithms, however, require large amounts of training data for accurate predictions. In the case of the LSM, this data includes landslide causative factors, as previously mentioned, and historical landslide data. However, landslides are a rare event in some regions, leading to the unavailability of extensive amounts of training data, which hinders the performance of ML algorithms. For a given area, landslide points (perceived as positive) are much less common than non-landslide points (perceived as negative), creating an imbalance between positive and negative points that predict affects.
Dr. Shukla’s team has developed a new ML algorithm that overcomes this issue of data imbalance for training the algorithm. It uses two under-sampling techniques, EasyEnsemble and BalanceCascade, to address the issue of unbalanced data in landslide mapping.
Data from landslides that occurred in the Mandakini river basin in the northwestern Himalayas, Uttarakhand, India, between 2004 and 2017 were used to train and validate the model. The results showed that the algorithm significantly improved the accuracy of LSMs, especially when compared to traditional machine learning techniques such as support vector machines and artificial neural networks.
“This new ML algorithm highlights the importance of data balancing in ML models and demonstrates the potential of new techniques for significant advances in the field,” said Dr. DP Shukla, Associate Professor in the School of Civil and Environmental Engineering. It also underscores the critical need for large amounts of data to train models accurately, especially in the case of geo-hazards and natural disasters where the stakes are high and human safety is at risk.
Dr Shukla believes that this study opens new avenues in the field of LSM and other geohazard mapping and management. It can be applied to other phenomena such as floods, avalanches, extreme weather events, rock glaciers and permafrost, thereby helping to reduce risks to human safety and property.
read all latest education news Here