Concerning the problems of the lack of standard words， fuzzy semantics and feature sparsity in news topic text， a news topic text classification method based on Bidirectional Encoder Representations from Transformers（BERT） and Feature Projection network（FPnet） was proposed. The method includes two implementation modes. In mode 1： the multiple-layer fully connected layer features were extracted from the output of news topic text at BERT model， and the final extracted text features were purified with the combination of feature projection method， thereby strengthening the classification effect. In mode 2， the feature projection network was fused in the hidden layer inside the BERT model for feature projection， so that the classification features were enhanced and purified through the hidden layer feature projection. Experimental results on Toutiao， Sohu News， THUCNews-L、THUCNews-S datasets show that the two above modes have better performance in accuracy and macro-averaging F1 value than baseline BERT method with the highest accuracy reached 86.96%， 86.17%， 94.40% and 93.73% respectively， which proves the feasibility and effectiveness of the proposed method.

Traditional machine learning methods fail to fully dig out semantic information and association information when classifying the sentiment polarity of online comment text. Although the existing deep learning methods can extract the semantic information and contextual information， the process is often one-way and there are some deficiencies in the process of obtaining the deep semantic information of comment text. Aiming at the above problems， a text sentiment analysis method was proposed by combining generalized autoregressive pretraining for language understanding model （XLNet） and RCNN （Recurrent Convolutional Neural Network）. Firstly， XLNet was used to represent the text features. And by introducing the segment-level recurrence mechanism and relative position information encoding， the contextual information of comment text was fully considered， thereby improving the expression ability of text features effectively. Then， RCNN was used to train the text features in both directions and extract the context semantic information of the text at a deeper level， thereby improving the comprehensive performance in the sentiment analysis task. The experiments with the proposed method were carried out on three public datasets weibo-100k， waimai-10k and ChnSentiCorp. The results show that the accuracy reaches 96.4%， 91.8% and 92.9% respectively， which proves the effectiveness of the proposed method in the sentiment analysis task.

Session-based recommendation aims to predict the user’s next click behavior based on the click sequence information of the current user’s anonymous session. Most of the existing methods realize recommendations by modeling the item information of the user’s session click sequence and learning the vector representation of the items. As a kind of coarse-grained information， the item category information can aggregate the items and can be used as an important supplement to the item information. Based on this， a Session-based Recommendation model of Multi-granular Graph Neural Network （SRMGNN） was proposed. Firstly， the embedded vector representations of items and item categories in the session sequence were obtained by using the Graph Neural Network （GNN）， and the attention information of users was captured by using the attention network. Then， the items and item category information given by different weight values of attention were fused and input into the Gated Recurrent Unit （GRU）. Finally， through GRU， the item time sequence information of the session sequence was learned， and the recommendation list was given. Experiments performed on the public Yoochoose dataset and Diginetica dataset verify the advantages of the proposed model with the addition of item category information， and show that the model has better effect compared with all the eight models such as Short-Term Attention/Memory Priority （STAMP）， Neural Attentive session-based RecomMendation （NARM）， GRU4REC on the evaluation indices Precision@20 and Mean Reciprocal Rank （MRR）@20.