基于学习者模型的教育资源推荐方法

Educational Resources Recommendation Method Based on Learner Model

为解决传统领域的推荐算法不完全适用于教育资源的问题,提出了一种基于学习者模型的混合推荐方法。首先,通过把神经网络中对文本信息的分类引入推荐算法中,缓解了冷启动问题;其次,通过对老用户的学习特征和学习能力进行类内计算,解决了传统协同过滤单一计算评分相似度和计算量大的问题。实验结果表明,该模型在教育资源大数据场景下取得了良好的效果。

In order to solve the problem that the traditional domain recommendation algorithm is not completely suitable for ed⁃ucational resources,a hybrid recommendation method based on the learner model is proposed.First,the cold start problem is allevi⁃ated by introducing the classification of text information in the neural network into the recommendation algorithm.Second,by per⁃forming in-class calculations on the learning characteristics and learning capabilities of old users,the traditional collaborative filter⁃ing single-computation score similarity and the problem of heavy calculation is solved.Experimental results show that the model has achieved good results in the context of big data in educational resources.