In this and the following chapters, we will apply the tree-based convolutional neural network (TBCNN) to the natural language processing. This chapter deals with constituency trees of natural language sentences, whereas the next chapter deals with dependency trees. In this chapter, we propose a constituency tree-based convolutional network (c-TBCNN). As usual, c-TBCNN can effectively extract structural information of constituency trees, which is aggregated in one or a few vectors for further information processing. c-TBCNN is applied in two sentence classification tasks: sentiment analysis and question classification. In both experiments, we achieve high performance similar to state-of-the-art models.
CITATION STYLE
Mou, L., & Jin, Z. (2018). TBCNN for constituency trees in natural language processing. In SpringerBriefs in Computer Science (pp. 59–71). Springer. https://doi.org/10.1007/978-981-13-1870-2_5
Mendeley helps you to discover research relevant for your work.