Department of Economics' Seminar Series - A talk by
Dr. Sushant K. Singh | Head of Artificial Intelligence Competency, Health, Insurance, and Life Sciences, New York City, NY, USA

Title
Application of Artificial Intelligence in Sustainable Arsenic Mitigation

Abstract
Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning have been applied to identify as well as to solve business, health, insurance, and life science challenges. However, their application in environmental problems is still in initial stages. This study aims to present the use of AI in various ecological challenges with a focus on sustainable arsenic mitigation. Data on socioeconomic, demographic, and other socio-behavioral factors of the community live in arsenic-contaminated areas were captured using a structured questionnaire. Six state-of-the-art ML algorithms, including Support Vector Machine (SVM), Decision Tree (DT), k-Nearest Neighbor (k-NN), Naïve Bayes (NB), Random Forests (RF), and Logistic Regression (LR) were used for developing arsenic awareness prediction models. Results suggest that the SVM and the RF ML models outclassed other ML models. Individuals' social network, caste, education, employment status, local social institutions, and NGOs were the most significant features across the models. The analysis also revealed that the association between the independent and dependent features was nonlinear therefore, nonlinear ML classifiers would be more suitable while developing prediction models on such data. ML could be more impactful in developing environmental models than traditional analytical methods, especially when the data is multidimensional. However, adequate sample size will be required for developing robust environmental models. Hybrid ML models could also be explored for such complex data.