Agricultural Question Classification Based on CNN of Cascade Word Vectors

Lei Chen, Jin Gao, Yuan Yuan, Li Wan

Abstract: Compared with traditional search engines, the query method of QA system is more intelligent and applicable in non-professional scenes, e.g., agricultural information retrieval. Question classification is an important issue in QA system. Since the particularities of agricultural questions, such as words sparsity, many technical terms, and so on, some existing methods are difficult to achieve the desired result in the agricultural question classification task. Hence, it is necessary to investigate how to extract as many useful information as possible from short agricultural questions to improve the efficiency of agricultural question classification. In order to solve this problem, the paper explores effective semantic representation of agricultural question sentences and proposes a method for agricultural question classification based on CNN of cascade word vectors. Different combinations of questions, answers, and synonym information are used to learn different cascade word vectors, which are taken as the input of CNN to construct the model of question classification. The experimental results show that our method can achieve better result in the agricultural question classification task.