This article explores the possibility of disaggregating query/question information in e-learning system online lectures or course recommendations. Information arrangement includes reading, parsing and classification of inquiry/question messages. Data extraction is a kind of shallow content processing. It finds a set of predefined applicable content in the feature language archives and performs common language processing through artificial intelligence strategies. During online lectures, many problems emerged in the listener’s minds, and the development of query optimization systems is of great significance to the evaluation of problems in online lectures. The results shows that our proposed method improve the classification of action verbs to a more accurate level. Later, we evaluated our proposed method, and measured a very high macro average for all one-sixth of the cognitive domain. We also provide the analytical examination in which we compared the designed method with the state of the art methods. The results indicate that the proposed method outperform the traditional methods

本文探讨了在电子学习系统在线讲座或课程推荐中分解查询/问题信息的可能性。信息安排包括查询，询问消息的阅读，解析和分类。数据提取是一种浅层内容处理。它在特征语言档案中找到一组预定义的适用内容，并通过人工智能策略执行通用语言处理。在线讲座期间，听者脑海中浮现出许多问题，查询优化系统的开发对在线讲座中问题的评估具有重要意义。结果表明，我们提出的方法将动作动词的分类提高到了更准确的水平。后来，我们评估了我们提出的方法，并为所有六分之一的认知域测量了非常高的宏观平均值。我们还提供了分析检查，其中我们将设计的方法与最先进的方法进行了比较。结果表明，该方法优于传统方法。