AI-ASSISTED BIODIVERSITY MONITORING AND SPATIAL ASSESSMENT FOR SUSTAINABLE WILDLIFE CONSERVATION AND COMMUNITY RESILIENCE IN TROPICAL FOREST ECOSYSTEMS

Abstract

Biodiversity documentation is essential for ecological monitoring, conservation planning, and environmental management. However, Samar State University (SSU) currently lacks a centralized and structured system for documenting faunal species, resulting in fragmented, inconsistent, and inaccessible biodiversity records. This study aimed to design and develop a Digital Photo-Based Faunal Information System that integrates mobile data collection, artificial intelligence-assisted species identification, cloud-based storage, and Geographic Information System (GIS) visualization. The study employed a developmental research design using Agile methodology, involving planning, design, development, testing, and evaluation phases. Data were collected through field observations, system implementation, and user evaluation using a structured questionnaire based on ISO/IEC 9126 software quality standards. The developed system enables users to capture faunal images, automatically record metadata, identify species using AI, and visualize geotagged observations on an interactive map. Evaluation results from 60 system users indicated high system performance, with composite mean scores of 4.44 (functionality), 4.39 (usability), 4.32 (reliability), and 4.27 (efficiency). The findings demonstrate that the system effectively improves biodiversity documentation, data accessibility, and ecological analysis. The system supports evidence-based conservation planning, environmental governance, and community resilience by providing accessible ecological information for sustainable resource management and decision-making. Integrating mobile computing, AI, GIS, and cloud technologies provide a scalable and reliable solution for institutional biodiversity management and conservation.