A Comprehensive Analysis of Marine Life Pollution Using Machine Learning Techniques on historical shipping Pollutants data

A Comprehensive Analysis of Marine Life Pollution Using Machine Learning Techniques on historical shipping Pollutants data.

  Dr. Santosh Kumar Singh 1, Anjali Singh2 and Srishti Dubey3   Amit Kumar Pandey4

1 H.O.D (IT), 2, 3, 4 PG Students

1,2,3,4 Department of Information Technology, Thakur College of Science and Commerce, Thakur Village, Kandivali (East), Mumbai-401107, Maharashtra, India

sksingh@tcsc.edu.inarpita9167@gmail.com.dsrishti381@gmail.com., amitpandey8089@gmail.com

Abstract. Oil spills significantly threaten marine and coastal ecosystems, resulting in devastating ecological and economic consequences. The current research highlights the key aspects surrounding oil spills, including their causes, environmental impacts, and mitigation strategies. Oil spills, resulting from accidents during offshore drilling, transportation, or natural seepage, pose severe ecological and societal challenges. These incidents release vast quantities of oil into marine ecosystems, leading to widespread environmental degradation. The impact of oil spills is influenced by factors such as spill volume, oil type, environmental conditions, and response effectiveness.  Ecologically, oil spills harm marine life through toxic effects, habitat destruction, and interference with reproductive cycles. Birds, marine mammals, fish, and shoreline organisms suffer from oil exposure, leading to long-term population declines. The economic consequences are significant, affecting fishers, tourism, and coastal industries. The clean-up process involves mechanical removal, chemical dispersants, and controlled burns, each with its environmental trade-offs.  Oil spill assessment methods encompass satellite monitoring, modeling, and ecological surveys to estimate damages and aid restoration efforts. Prevention measures include stricter regulations, technological advancements, and industry best practices. Public awareness and international cooperation are vital for enhancing spill preparedness and response.  In the paper, oil spills continue to challenge environmental sustainability and economic stability. Addressing their multifaceted impacts demands an integrated approach involving mitigation, rigorous assessment, and global collaboration to prevent future disasters and safeguard our oceans and coastlines. An expansive dataset gathered from Queensland, Australia, incorporating intricate information about ships, regions, longitudes, latitudes, and pollutants, the study markedly advances the capability to identify oceanic oil spills and enhances our understanding of the resulting impact on marine life. In that, algorithms are used of machine learning techniques such as SVC and LogisticRegression. In the course of the study, machine learning techniques are applied, featuring algorithms like Support Vector Classification (SVC) and Logistic Regression. These algorithms play a crucial role in the analysis, contributing to the extraction of valuable insights from the dataset.

Keywords: Bilge, Diesel, Hydraulic Oil, Types of Ship, Machine learning algorithms