Artificial Intelligence Analysis of Sacred Groves as Refuges for Animal Diversity

Abstract

Introduction: Sacred forests are critical sites where the animal diversity is protected through providing them stable habitats and protection from human disturbances. Sacred forests have been practicing traditional methods of conservation, which help in maintaining biodiversity. However, analysis of animal diversity patterns needs advanced scientific methods. In this research, use artificial intelligence methods to assess whether sacred forests are appropriate refuges for animal diversity. Materials and Methods: The ecological analysis system was constructed through an artificial intelligence approach by using 1,200 occurrence data of animals with 10 ecological features such as vegetation, tree canopy cover, diversity of trees, temperature, humidity, habitat disturbance, and human disturbance. 70% of data was used as a training sample while 30% of data was used as a testing sample. Machine learning approaches including SVM (Support Vector Machine), ANN (Artificial Neural Network), and RF (Random Forest) were employed for analysis of species-habitat relationships. Results and Discussion: According to the results obtained from the experiments, it is clear that the AI models have successfully managed to capture patterns affecting the diversity of the animals. The Random Forest model was successful with 95.6% accuracy, 94.8% precision, 95.2% recall, 95.0% F1-Score, and AUC score of 0.97 while the next best algorithm was ANN which had an AUC score of 0.93 and SVM which had an AUC score of 0.88. Conclusion: The paper brings to light the use of AI in assessing and conserving biodiversity. AI in ecological evaluation could aid in monitoring the sacred groves, determine the suitability of the habitats and enhance efforts to conserve animal biodiversity