Scholarlink

This study presents a hybrid recommendation system that integrates Machine Learning (ML) and Generative Artificial Intelligence (AI) to predict and mitigate corrosion-related degradation in metallic elements of Underwater Cultural Heritage caused by environmental factors such as temperature, salinity, and pressure. Real-time oceanographic data from SeaDataNet CDI and ODATIS-Coriolis is analyzed using multiple linear regression, achieving high predictive accuracy (R \(^{2}\) = 0.864). The results show that temperature and salinity have a strong influence on corrosion, while pressure has a lower effect. The system consists of three layers: (1) Data acquisition and processing, (2) ML-based corrosion prediction, and (3) The integration of AI-powered semantic recommendation module for mitigation strategies; which simulates environmental conditions and suggests preventive measures. By combining predictive analysis with adaptive AI-based recommendations, this approach enhances conservation planning for submerged archaeological sites.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Recommendation System for the Degradation of Underwater Cultural Heritage Based on Machine Learning and Generative Artificial Intelligence

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Juan M. Núñez V.,
  • Carolina Villoria-Torres,
  • Marta Plaza-Hernández,
  • Javier Prieto-Tejedor

摘要

This study presents a hybrid recommendation system that integrates Machine Learning (ML) and Generative Artificial Intelligence (AI) to predict and mitigate corrosion-related degradation in metallic elements of Underwater Cultural Heritage caused by environmental factors such as temperature, salinity, and pressure. Real-time oceanographic data from SeaDataNet CDI and ODATIS-Coriolis is analyzed using multiple linear regression, achieving high predictive accuracy (R \(^{2}\) = 0.864). The results show that temperature and salinity have a strong influence on corrosion, while pressure has a lower effect. The system consists of three layers: (1) Data acquisition and processing, (2) ML-based corrosion prediction, and (3) The integration of AI-powered semantic recommendation module for mitigation strategies; which simulates environmental conditions and suggests preventive measures. By combining predictive analysis with adaptive AI-based recommendations, this approach enhances conservation planning for submerged archaeological sites.