Application of machine learning and neural network technology in art design

In the digital art domain, the integration of intelligent design and analytical capabilities necessitates effective methods for automatically discerning and evaluating artworks. This research suggests a machine learning-based neural network method to the challenge. To investigate emotional resonance in numerous art forms across disciplines, a deep recurrent neural network is built. A new cross-domain edge cloud model uses cloud computing advances. This architecture offloads streaming media services to edge network sub-clouds, revolutionising storage and compute. Edge networks make cross-media data collecting easy, enabling analysis. Deep neural networks analyse visual and linguistic input to classify viewer emotions via multimodal classification. Experimental results show that the model can accurately identify unlabeled cross-media data. The technique also mitigates the possibility of erroneous emotion representation in AI systems by addressing artificial emotion simulation. The MMBT model outperformed others with 66.33% accuracy and 62.24% F1 value. This research provides a complete framework for discovering emotional nuances in cross-media art and intelligent art design and analysis.