Leveraging Spectral Clustering and Long Short-Term Memory Techniques for Green Hotel Recommendations in Saudi Arabia DOI Open Access
Abdullah Alghamdi

Sustainability, Journal Year: 2025, Volume and Issue: 17(5), P. 2328 - 2328

Published: March 6, 2025

Online recommendation agents have demonstrated their value in various contexts by helping users navigate information overload, supporting decision-making, and influencing user behavior. There is a lack of studies focusing on systems for green hotels that utilize user-generated content from social networking e-commerce platforms. While numerous explored the use real-world datasets hotel recommendations, development specifically remains underexplored, particularly context Saudi Arabia. This study attempts to develop new approach recommendations using text mining Long Short-Term Memory techniques. Latent Dirichlet Allocation used identify main aspects users’ preferences content, which will help recommender system provide more accurate users. preference prediction based numerical ratings. To better perform clustering technique overcome scalability issue proposed system, when there large amount data datasets. Specifically, spectral algorithm cluster ratings hotels. evaluate method, 4684 reviews were collected Arabia’s TripAdvisor platform. The method was evaluated its effectiveness solving sparsity issues, accuracy, scalability. It found predicts customers’ overall comparison results provides highest precision (Precision at Top @5 = 89.44, Precision @7 88.21) lowest error (Mean Absolute Error 0.84) recommendations. author discusses presents research implications findings method.

Language: Английский

Leveraging Spectral Clustering and Long Short-Term Memory Techniques for Green Hotel Recommendations in Saudi Arabia DOI Open Access
Abdullah Alghamdi

Sustainability, Journal Year: 2025, Volume and Issue: 17(5), P. 2328 - 2328

Published: March 6, 2025

Online recommendation agents have demonstrated their value in various contexts by helping users navigate information overload, supporting decision-making, and influencing user behavior. There is a lack of studies focusing on systems for green hotels that utilize user-generated content from social networking e-commerce platforms. While numerous explored the use real-world datasets hotel recommendations, development specifically remains underexplored, particularly context Saudi Arabia. This study attempts to develop new approach recommendations using text mining Long Short-Term Memory techniques. Latent Dirichlet Allocation used identify main aspects users’ preferences content, which will help recommender system provide more accurate users. preference prediction based numerical ratings. To better perform clustering technique overcome scalability issue proposed system, when there large amount data datasets. Specifically, spectral algorithm cluster ratings hotels. evaluate method, 4684 reviews were collected Arabia’s TripAdvisor platform. The method was evaluated its effectiveness solving sparsity issues, accuracy, scalability. It found predicts customers’ overall comparison results provides highest precision (Precision at Top @5 = 89.44, Precision @7 88.21) lowest error (Mean Absolute Error 0.84) recommendations. author discusses presents research implications findings method.

Language: Английский

Citations

0