【摘要】 随着"互联网+"时代的来临,在线教育平台推荐系统在大数据技术的帮助下,相较于传统教育有着显著的优势。针对数据稀疏性的教育平台的推荐算法,对推荐系统数据的稀疏性问题及架构进行了分析,并对基于二部图的推荐算法进行了优化,最后对基于改进二部图的推荐算法进行了测试分析。结果显示,虽然二部图优化算法稳定性尚不足,但在整体推荐效果上具有一定的优势,二部图优化算法与协同过滤算法相比,在准确率和召回率上的优势,分别要高25%和23%。更多还原

【Abstract】 With the advent of the "Internet plus" era, the online education platform recommendation system has great advantages over traditional education with the help of big data technology. This research aims at the recommendation algorithm of the education platform with sparse data, analyzes the data sparsity problem and architecture of the recommendation system, optimizes the recommendation algorithm based on bipartite graph, and finally tests and analyzes the recommendation algorithm based on improved bipartite graph. The results show that although the stability of the bipartite graph optimization algorithm is not enough, it has certain advantages in the overall recommendation effect. Compared with the collaborative filtering algorithm, the bipartite graph optimization algorithm has 25% higher accuracy and 23% higher recall, respectively.更多还原