Prediction on Bird Diversity from Datasets of Rural and Industrial Area of Asansol, West Bengal: A Machine Learning Approach

Debdyuti Sengupta, Soumendra Nath Talapatra

Abstract: The objective of the present study was to predict the accuracy of datasets of bird diversity in rural and industrial area of Asansol, West Bengal during monsoon season. This study was performed through machine learning (ML) algorithms such as BayesNet (BN), NaiveBayes (NB), logistic regression (LR) and Random Tree (RT) by using WEKA tool, version 3.8.6. The study was separately based on 3 attributes such as Order, Numbers, and Effects (High, Moderate and Low) to know overall prediction accuracy of dataset as per 10-fold cross validation (CV) test for each algorithm. In the present study, the weighted average of precision recall curve (PRC) values obtained 100.0% for rural and industrial area on these algorithms. It is concluded that these ML algorithms especially BN and RT predicted accurately from the dataset and obtained rich information with statistical interpretation, which confirmed lower diversity of avifauna.

Keywords: Asansol, Bird diversity, Machine learning algorithms, Industrial area, Prediction accuracy of dataset, Rural area

How to Cite?: Debdyuti Sengupta, Soumendra Nath Talapatra, "Prediction on Bird Diversity from Datasets of Rural and Industrial Area of Asansol, West Bengal: A Machine Learning Approach", Volume 14 Issue 7, July 2025, International Journal of Science and Research (IJSR), Pages: 466-469, https://www.ijsr.net/getabstract.php?paperid=SR25707002733, DOI: https://dx.doi.org/10.21275/SR25707002733