景区特色智能推荐模型

基于协同过滤与内容推荐融合算法，构建 “用户特征 - 场景标签 - 文化元素” 关联规则：输入地域 / 口味时，通过自然语言处理技术解析饮食文化关键词，关联地方特色美食数据库；输入偏好 / 天气时，结合非遗文化资源库与路径规划算法，生成个性化非遗体验路线；对用户消费记录、会员等级等数据进行去标识化处理（剔除姓名、手机号等敏感信息），仅保留行为标签用于模型训练。

Based on the hybrid algorithm combining collaborative filtering and content-based recommendation, we construct association rules structured as "user characteristics - scene tags - cultural elements". When receiving region or flavor inputs, parse dietary culture-related keywords via natural language processing (NLP) techniques and link to the local specialty food database; When receiving preference or weather inputs, combine the intangible cultural heritage (ICH) resource repository and path planning algorithms to generate personalized intangible cultural heritage experience routes; Perform de-identification processing on data such as user consumption records and membership levels, removing sensitive information including names and mobile phone numbers, and only retain behavior tags for model training.