基于多维数据融合的个性化旅游信息推荐模型研究  

Research on Personalized Tourism Information Recommendation Model Based on Multi-dimensional Data Fusion

作　　者：吕春丽 LYU Chunli(Chongqing Vocational Institute of Tourism,Chongqing 409000,China)

机构地区：[1]重庆旅游职业学院,重庆409000

出　　处：《移动信息》2024年第10期320-322,326,共4页Mobile Information

基　　金：2021年重庆市教委科技项目(KJQN202104602);2023年重庆市教委科技项目(KJQN202304606、KJQN202304608)。

摘　　要：为了更好地促进旅游信息个性化推荐效果的提升,需对用户行为信息进行深度挖掘。文中基于多维数据融合,构建并研究了个性化旅游信息推荐模型。首先,结合用户的历史浏览、检索、点赞及评论数据,分析用户对不同景区的兴趣度。其次,结合用户的旅行次数、旅行时间以及具体的旅行地点信息,分析用户的偏好选择对旅游大数据的影响力。接着,结合用户对景区兴趣度的分析结果,综合计算用户对景区所在地的偏好选择。最后,以地区偏好为范围,将对旅游大数据影响力较高用户的兴趣度TOP-N景区信息作为最终的推荐结果。测试结果表明,归一化折损累计增益NDCG随着推荐信息数量k的增加呈现出稳定提升的趋势,且相比对比模型,文中提出的模型得到的参数值始终处于较高水平,说明该模型具有可靠的推荐效果。In order to better promote the improvement of personalized recommendations for tourism information,it is necessary to conduct in-depth mining of user behavior information.Based on multidimensional data fusion,a personalized tourism information recommendation model was constructed and studied in the paper.Firstly,based on the user's historical browsing,retrieval,liking,and commenting data,analyze the user's interest in different scenic spots.Secondly,based on the user's travel frequency,travel time,and specific travel location information,analyze the influence of user preferences on tourism big data.Next,based on the analysis of users̓interest in the scenic area,comprehensively calculate users̓preference choices for the location of the scenic area.Finally,based on regional preferences,the TOP-N scenic spot information of users with high influence on tourism big data will be used as the final recommendation result.The test results show that the normalized cumulative loss gain shows a stable improvement trend with the increase of the recommended information quantity k.Compared with the comparative model,the parameter values obtained by the proposed model in the paper are always at a high level,indicating that the model has reliable recommendation performance.

关 键 词：多维数据融合 个性化旅游信息推荐模型 景区兴趣度 大数据影响力 所在地偏好 归一化折损累计增益 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]