AstroExpert: A One-Stop Platform for AI-Driven Astronomical Observation and Smart Commerce

The "Astro Expert" platform successfully addresses the inherent difficulties of amateur observation planning by bridging the gap between raw meteorological data and user-friendly web interfaces. By deploying a specialized architecture featuring thirty isolated TensorFlow models, the project achieved highly accurate, threshold-bound stargazing predictions (88.2% accuracy). The seamless integration of a domain-restricted Gemini chatbot, live celestial tracking, and a tailored e-commerce ecosystem successfully centralizes the astronomical planning process into one cohesive, secure tool. Future Scope: Future iterations of this project will aim to transition the 30-model setup into a single, continuous-learning network (such as an LSTM) capable of autonomously determining complex weather thresholds, thereby reducing computational overhead. To mitigate reliance on external APIs, future development will include local caching databases for offline functionality and the deployment of a compressed, edge-hosted Large Language Model (LLM) to replace cloud-dependent chatbot services, ensuring a fully self-contained ecosystem.

[1] Mislav Jovanić (2024): Application of Artificial Intelligence in the Creation of Web Content. AI-Driven Content Generation to automatically create complex web components based on simple inputs. [2] Muhammet Sait Ag (2025): Websites Quality and Online Success: A Bibliometric Analysis. Bibliometric Analysis of existing gear reviews and success rates to inform e-commerce categorization. [3] Gulmira Bekmanova (2024): Requirements for the Development of a Website Builder with Adaptive Design. Implementation of Server-Side Rendering to ensure fast, responsive, and SEO-optimized delivery. [4] Sainu Nannat (2024): Designing Professional Websites with Odoo Website Builder. A practical guide to crafting elegant, intuitive websites using operational tools and builders. [5] Zeynep Aydınlı (2025): AI-Supported and Interactive Exploration, Information and Education of Space. Overcoming traditional teaching limitations by integrating modern technologies into space education. [6] Shuvam Chakraborty: LOCI: Learning Low Overhead Collaborative Interference Cancellation for Radio Astronomy (2023) [7] L Cabayol: The PAU survey: estimating galaxy photometry with deep learning. (2021) [8] Noble Kennamer: Active learning with RESSPECT: Resource allocation for extragalactic astronomical transients (2020)