APPLICATION OF ADVANCED MACHINE LEARNING AND DATA ANALYTICS TO ENHANCE ESG PERFORMANCE: A CASE STUDY OF THE SUSTAINABLE OIL PALM INDUSTRY IN NIGERIA

The efficient assessment of Environmental, Social, and Governance (ESG) performance is critical for achievingsustainability, and the global goal of net-zero emissions by 2050. Traditional ESG scoring methods may be effective in varioussectors, but are less applicable to the oil palm industry due to their inherent subjectivity. This paper addresses these limitations, byapplying a hybrid approach through the use of Machine Learning (ML) and data analytics, to evaluate ESG Performance Indicatorsin Nigeria's sustainable agriculture sector, specifically within the oil palm industry. The objective is to identify areas ofimprovement, and reduce the carbon footprint during the oil palm plantation production phase. The study employs a comprehensivedata-driven methodology, including Data Collection, Keyword Analysis, Clustering, Topic Modeling, Term Frequency-InverseDocument Frequency (TF-IDF) Analysis, Sentiment Analysis, Principal Component Analysis, and Advanced Clustering. Thesetechniques analyze ESG Performance Indicators, focusing on sustainability, compliance, risk, and safety. The results show thatadvanced analytical methods, can effectively assess ESG performance. Keyword Analysis and Clustering uncover key themes andpatterns, while Sentiment Analysis reveals that the farm's ESG practices are generally positive, but highlight the need for improvedsustainability measures. The developed ESG scoring system, quantitatively identifies strengths, and areas for improvement, enablingtargeted interventions. In conclusion, this hybrid analytical approach proves optimal, for ESG performance assessment in the oilpalm industry, and is adaptable for similar evaluations across various sectors globally.