Computational mining of empirical literature on forest recreation: A semantic-driven topic modeling approach based on advanced contextual embeddings

This paper conducts a thematic literature review of advancements in forest recreation research. Methodologically, we analyzed a corpus of empirical articles in economics and management published between 2015 and 2024, sourced from the Scopus and Web of Science databases. Using the BERTopic algorithm for semantic topic modeling, we identified six distinct topics, each highlighting specific aspects and geographical diversity in forest recreation. These topics were evaluated using a thematic map. The findings reveal that (Economic value, planning, management, and visitor preferences in forest recreation) is an influential topic, significantly shaping recent discussions and practices. Additionally, (Economic valuation, injury risks, and visitor preferences in forest recreation) and (Visitor preferences, digital recreation, and heritage value in forest recreation) are well-developed themes, offering a robust research base and deep insights into key questions. On the other hand, (Visitor preferences and willingness to pay for forest recreation improvements) represents an emerging topic, gaining traction and likely to become a major focus. Meanwhile, (Economic, ecological capacity, and visitor behavior in forest recreation and management) and (Urban forests, visitor behavior, and management innovations) are cross-cutting topics, spanning multiple interconnected fields and providing a broader perspective. This review seeks to inform policymakers, researchers, and forest managers by highlighting recent advancements in international forest recreation over the past decade, a period characterized by significant environmental changes and strategic adjustments crucial for sustainable forest resource management.