**3. Hybrid Deep-Learning Model**

The whole process of classification is shown in Figure 1. This chapter will describe each layer in the subsections and combine them as the BLSTM-C model at the end of this chapter.

**Figure 1.** Classification process.

#### *3.1. Input Layer*

The first layer is the input layer that does some processing on the input. This layer can be divided into two parts: remove the stop words and segmentation. The first step removes the stop words from the article so that the information in the text will be more concentrated. The second step is a unique processing for Chinese language. Since Chinese words do not have a space to separate them, it would be necessary to segment them into words manually so that we can represent each word as vector.
