User Profiling for CSDN: Keyphrase Extraction, User Tagging and User Growth Value Prediction

Six tables, five figures and one algorithm of this paper. Table 1 shows candidate phrases. Table 2 demonstrates performance of different types of features. Table 3 summarizes notations used in this paper. Table 4 shows the statistics of data sets. Table 5 compares scores of methods. Table 6 shows performance of models. Algorithm 1 shows semi-supervised regression algorithm. Figure 1 shows candidate phrase generation procedure. Figure 2 is the framework of a two-phase model. Figure 3 depicts the architecture of the neural network. Figure 4 shows the architecture of user growth value prediction. Figure 5 presents bagging of primary model.