<![CDATA[The monitoring of Near-Earth Objects (NEO) faces a gap between complex NASA raw data availability and public understanding of asteroid hazards. This study bridges pure astronomy and Information Systems Engineering by developing SINEO (Near-Earth Objects Information System), an integrated web platform using Laravel 12. SINEO automatically connects to the NASA CNEOS API to translate complex astronomical data into precise, real-time, and communicative information. Developed using the Scrum methodology, the system adopts a hybrid architecture combining PHP (Laravel) for the interface and Python (TensorFlow) for the artificial intelligence engine. The AI module employs a Dual-Stacked Recurrent Neural Network (RNN) model with Recursive Multi-step Forecasting. System functionality was validated through Black Box Testing, while user satisfaction was evaluated using the System Usability Scale (SUS). Results show the API integration successfully automates data synchronization without rate-limiting issues. The RNN model training over 100 epochs achieved an optimal Mean Squared Error (MSE) of 0.000537. Using Root Mean Squared Error (RMSE) calibrated with a 3.5 variance scale, the model yielded a 91.9% convergence accuracy in normalized space, effectively forecasting asteroid miss distance for 7 days ahead without overfitting bias. The dashboard visualizes automated risk assessments via a Traffic Light System (TLS) integrating Potentially Hazardous Asteroid (PHA) status. Usability evaluation with 30 respondents resulted in an average SUS score of 79.50 (Acceptable, Grade B), proving SINEO is highly effective, efficient, and satisfying for users.]]>
