基于NLP的大学生自主学习智能问答系统设计

Design of intelligent question answering system for self-directed learning of college students based on NLP

自主学习能力作为高校大学生重要的一种学习能力,已成为大学生应当具备的一项重要的基本素质。大学生自主学习最大难题是遇到问题时不能实时、精准、有效地解决。针对这个问题,提出基于自然语言处理(natural language processing,NLP)的自主学习智能问答系统。系统使用基于转换器的双向编码表征预训练模型提取输入序列的表示,然后通过注意力机制将表示传递到下游神经机器翻译模型的编码器和解码器的每一层中,最后联结成自主学习智能问答模型。在开源的中文医学问答数据集——cMedQA2上进行训练和测试自主学习智能问答模型。实验结果表明,与其它模型相比,该模型的自然语言处理评价指标BLUE值有明显提高。进一步分析预测样例,对于同一问题,预测值的表达基本与真实值相符。

Self-directed learning ability,as an important learning ability of college students,has become an important basic quality that college students should possess.The biggest problem of self-directed learning for college students is that they cannot be solved in real time,accurately and effectively when they encounter questions.To solve this problem,a self-directed learning intelligent question answering system is proposed based on natural language processing(NLP).The system uses the bidirectional encoding representation from transformers(BERT)pre-training model to extract the representation of the input sequence,and then transmits the representation to each layer of the encoder and decoder of the downstream neural machine translation(NMT)model through the attention mechanism,and finally connects to the self-directed learning intelligent question answering system model.And the model is trained and tested on the open source Chinese medical question and answer data set cMedQA2.The experimental results show that compared with other models,the BLUE value of the NLP evaluation index of this model is significantly improved.Further analysis of the prediction example shows that for the same problem,the expression of the predicted value is basically consistent with the true value.