基于知识图谱的多源无标签数据推荐算法  

Multi-Source Unlabeled Data Recommendation Algorithm Based on Knowledge Map

作　　者：葛湘巍 刘洋 GE Xiang-wei;LIU Yang(College of Computer Science and Technology,Inner Mongolia Normal University,Hohhot Inner Mongolia 010022,China)

机构地区：[1]内蒙古师范大学计算机科学技术学院,内蒙古呼和浩特010022

出　　处：《计算机仿真》2024年第10期473-477,共5页Computer Simulation

基　　金：内蒙古师范大学校级(2022JBQN109);内蒙古高等学校科学技术研究项目(NJZY20020)。

摘　　要：为了全面提升多源无标签数据的推荐能力,实现高效准确的数据推荐效果,提出一种基于知识图谱的多源无标签数据推荐算法。构建多源无标签数据知识图谱,向量化处理多源无标签数据。融合数据的知识表示和时间编码,经过映射处理获取深层次的用户行为项特征。通过Transformer机制编码用户语义表示和解码用户偏好,对用户语义偏好和目标内容展开预测,获取用户和候选多源无标签数据发生交互的概率,实现多源无标签数据推荐。实验结果表明,所提算法可以快速为用户提供准确的多源无标签数据推荐结果。In order to comprehensively improve the recommendation ability of multi-source unlabeled data and achieve efficient and accurate data recommendation,a recommendation algorithm for multi-source unlabeled data based on knowledge map was presented.Firstly,we built a knowledge map of multi-source unlabeled data,and vectorized the multi-source unlabeled data.After fusing the knowledge expression and time coding of data,we obtained the indepth user behavior item features.Then,we used the Transformer mechanism to encode user semantic representation and decode user preferences.Meanwhile,we predicted user semantic preferences and target content,thus obtaining the probability of interaction between users and candidate multi-source unlabeled data.Finally,we realized the multisource unlabeled data recommendation.Experimental results show that the proposed algorithm can quickly provide users with accurate recommendation results of multi-source unlabeled data.

关 键 词：知识图谱 多源无标签数据 语义偏好 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]