Design of a context-aware AR smart guiding platform for enhancing cultural heritage experience in rural tourism

Abstract

In recent years, the incorporation of developing technologies like Augmented Reality (AR) has altered how people interact with cultural assets. It describes the design and development of a Context-Aware AR Smart Guiding Platform (CA-ARSG) to improve the cultural heritage experience in rural tourism settings. The proposed system leverages deep learning techniques to intelligently personalize AR content based on user preferences, environmental conditions, and real-time contextual data, including location, weather, and site popularity. Data preprocessing involves handling missing values and applying Z-score normalization to ensure clean and standardized input for accurate model predictions. Specifically, an Efficient Gannet Optimized Attention Convolutional Long Short-Term Memory (EGO-Att-ConLSTM) model is employed to analyze user travel history and predict interest patterns. The platform utilizes mobile AR to overlay digital content such as 3D reconstructions, interactive narratives, and historical insights on physical cultural sites in rural areas. A context management module dynamically adjusts AR experiences based on local relevance and user behavior. The system was evaluated through usability studies and performance testing in selected rural heritage sites. Results indicate that the EGO-Att-ConLSTM model improved user engagement, satisfaction, and learning outcomes, with the deep learning-enhanced recommendation engine outperforming baseline models in accuracy by 96.2%. It contributes to the advancement of intelligent tourism technologies by demonstrating how context-aware AR combined with deep learning can significantly enrich rural cultural experiences while supporting sustainable heritage promotion.

Graphical Abstract