Interactive Genetic Algorithm with Convolutional Neural Network-based Filtering for Personalized Interior Space Design

This research presents an Interactive Genetic Algorithm (IGA) system that merges human intuition with computational intelligence to revolutionize interior space design. Acting as a collaborative partner, it learns from user feedback to evolve layouts in real time, integrating 3D spatial data (floor plans, furniture catalogs) and style preferences (minimalist, industrial, etc.). A hybrid CNN-IGA architecture filters impractical options early, minimizing decision fatigue while encouraging creativity. The system achieves 97.6% accuracy in matching preferences and 95.3% recall for unique ideas, outperforming ANN, SVM, and standard GA. Unlike static algorithms, it translates subjective concepts like "cozy" into measurable design principles through iterative refinement. Beyond metrics, it redefines the designer-AI relationship—enhancing, not replacing, human judgment. This dynamic toolkit benefits professionals and homeowners alike, with potential applications in VR staging and sustainable optimization. Ultimately, the system proves that the best designs emerge when technology and creativity collaborate.