基于日志处理的植保作业推荐算法研究

Research on Plant-Protecting Operation RecommendationAlgorithm Based on LogData Processing

摘要: 本文提出了一种基于日志处理的植保作业推荐方法，利用Storm框架技术搭建了一个日志实时分析计算系统，设计并开发了混合用户特征和协同过滤的植保作业推荐算法与基于内容过滤的植保作业推荐算法。混合用户特征和协同过滤的植保作业推荐算法首先计算与目标用户最相似的邻居用户，按其作业加权评分结果获取适宜的植保作业推荐给无人机用户；基于内容过滤的植保作业推荐算法基于用户的搜索与查看记录，为用户推荐可能感兴趣的植保作业。系统测试与分析表明，基于日志处理的植保作业推荐算法能够按设定规则为用户推荐植保作业，提高了系统的智能性、便捷性和可用性。

Abstract: The scientific and reasonable recommendation algorithm for plant-protecting operation can effectively assure the safety of UAV (Unmanned Aerial Vehicles), as well improve the efficiency of plant protecting work and its management level. This paper proposed a method of operation recommendation for plant-protecting UAV users based on log data processing, built a log data real-time analysis and calculation system using Storm framework technology, and developed an operation recommendation algorithm combining user feature and collaborative filtering and another plant-protecting operation recommendation algorithm based on content filtering. In the former algorithm, the most similar neighbor users of the target user were investigated and calculated, and then the appropriate operation recommendation list was obtained according to the calculation result of their weighted scores. The later recommendation algorithm was based on content filtering which recommended the potential plant-protecting operations for the corresponding user by analyzing his search and browser history and records. System run and analysis showed that the log data processing based algorithm could give users the best plant-protecting operation recommendations according to the setup rules, which could potentially improve the system’s intellectuality, convenience and usability.