The authors propose a method to extract individual travel spatiotemporal behaviors from social media data, and then mine the group based on massive spatiotemporal behaviors. This study collects more than 40 million global geographic microblogs from users who have visited Suzhou, extracts 88270 tourism trajectories, and identifies 36 classes of inter-city tourism movement patterns in five categories. It is found that the extracted patterns conform to the LCF theoretical model; besides the simple movement patterns, there are more complex composite movement patterns. Based on big data, more accurate tourism movement patterns can be obtained, which helps tourism managers understand tourists’ trends and preferences, adjust destination marketing strategies, optimize tourism resource allocation, and provide better services for tourists.

提出一种从社交媒体大数据中提取个体旅游时空行为, 再基于海量旅游时空行为挖掘群体城市间移动模式的方法。采集4000多万条到访过苏州市用户的全球地理微博, 从中提取88270条旅游时空行为轨迹, 识别出5类36种城市间旅游移动模式。结果表明, 提取的移动模式符合LCF理论模型; 除简单移动模式外, 还存在更加复杂的复合移动模式。基于大数据能够得到更全面更精准的旅游移动模式, 有助于旅游管理者了解游客动向及偏好, 调整目的地营销策略, 优化旅游资源配置, 为游客提供更好的服务。