基于用户情景推断的军事信息Hybrid-CF推荐算法  

Hybrid-CF Recommendation Method for Military Information Based on User Scenario Inference

作　　者：陶飞飞[1] 陈诚[1] 石敏芳 邓劲柏 TAO Feifei;CHEN Cheng;SHI Minfang;DENG Jinbai(School of Computer and Software,Hohai University,Nanjing 210098,China)

机构地区：[1]河海大学计算机与软件学院,南京210098

出　　处：《计算机测量与控制》2025年第4期186-191,共6页Computer Measurement &Control

基　　金：国家自然科学基金项目(62302151)。

摘　　要：针对当前军事信息推荐方法未对用户所处时间、地理位置和用户场景做区分,以及未考虑军事用户的信息需求与所处场景的关联性,导致推荐结果固定、单一化的问题,设计了基于用户情景推断的军事信息Hybrid-CF(混合协同过滤)推荐方法;融入军事用户情景要素对传统协同过滤算法进行了改进,通过计算当前情景信息与历史信息的相似度,更加准确地推断出当前军事用户的所处情景,继而给军事用户推荐符合其需求的特定情景下的军事信息;为了解决推荐算法矩阵稀疏、效果单一等问题,引入了加权平衡因子将不同的推荐算法进行动态加权得到融合情景信息的Hybrid-CF推荐算法,并通过控制因子λ对加权平衡因子进行动态调整;实验结果表明,所提出的Hybrid-CF推荐算法在准确率和召回率上均体现了良好的提升效果。Aiming at the problem that the current military recommendation algorithm does not distinguish between the user’s time,geographical location and user scenario,and the disregard of the relevance of military users’information needs and scenario,resulting in fixed and single recommendation results,this paper designs a military information hybrid collaboration filter(Hybrid-CF)recommendation method based on user scenario inference.The scenario information of military user is integrated to improve the collaborative filtering algorithm,and calculate the similarity between the current scenario and historical information,so as to accurately infer the scenario of military user and provide the recommendation service for military user in a specific scenario.In order to solve the problems of sparse matrix and single effect of the recommendation algorithm,the weighted balance factor is introduced to dynamically weight different recommendation algorithms and achieve the Hybrid-CF recommendation algorithm that integrates the fusion scenario information,dynamically adjusting the weighted balance factor through the control factorλ.Experimental results show that the proposed Hybrid-CF recommendation algorithm has good peformance in improving accuracy and recall rate.

关 键 词：协同过滤 融合情景 情景推断 加权平衡因子 混合推荐 

分 类 号：TP391.3[自动化与计算机技术—计算机应用技术]