<![CDATA[Recent advancements in mobile technology and machine learning have enabled the development of practical tools, such as Android applications, to assist in real-time fish species identification, particularly in the context of freshwater fisheries in Indonesia. Objective: This research aims to design and implement an Android application that helps anglers accurately identify and categorize freshwater fish species native to Indonesia. The app integrates machine learning-based image recognition to provide a practical tool for fishing enthusiasts while supporting conservation efforts for Indonesia’s freshwater biodiversity. Methodology: A quantitative approach was employed, focusing on mobile application development using Kotlin for Android. The application uses a TensorFlow Lite-based image recognition model for real-time image processing on mobile devices. Data for the model were gathered from publicly available fish species datasets. The system was tested across multiple Android devices to evaluate compatibility and efficiency. Findings: The application successfully identifies and classifies various freshwater fish species in Indonesia, providing users with accurate species profiles, biological characteristics, and appropriate bait recommendations. The system operates efficiently in real-time on mobile devices without relying on cloud computing, ensuring accessibility in remote areas. Testing results across different Android devices confirm the app's robustness and user-friendly interface. Implications: This research demonstrates the integration of mobile technology and machine learning in fisheries, offering a valuable tool for both recreational and professional anglers. The app promotes awareness of freshwater fish species preservation and supports sustainable fishing practices. Additionally, it can serve educational purposes by enhancing knowledge of local biodiversity and fostering fish conservation efforts. Originality: This research introduces an innovative mobile-based solution to freshwater fish identification. Unlike previous studies, which focused on desktop-based methods, this study offers a practical mobile application that operates efficiently in real-time on-site. The originality lies in combining machine learning and mobile technology to address fish identification challenges while contributing to biodiversity conservation.]]>
