Previous research on sentiment analysis mainly focuses on binary or ternary sentiment analysis in monolingual texts. However, in today's social media such as micro-blogs, emotions are often expressed in bilingual or multilingual text called code-switching text, and people's emotions are complex, including happiness, sadness, angry, afraid, surprise, etc. Different emotions may exist together, and the proportion of each emotion in the code-switching text is often unbalanced. Inspired by the recently proposed BERT model, we investigate how to fine-tune BERT for multi-label sentiment analysis in codeswitching text in this paper. Our investigation includes the selection of pre-trained models and the finetuning methods of BERT on this task. To deal with the problem of the unbalanced distribution of emotions, a method based on data augmentation, undersampling and ensemble learning is proposed to get balanced samples and train different multi-label BERT classifiers. Our model combines the prediction of each classifier to get the final outputs. The experiment on the dataset of NLPCC 2018 shared task 1 shows the effectiveness of our model for the unbalanced code-switching text. The F1-Score of our model is higher than many previous models.

Risperidone is routinely used in the clinical management of schizophrenia, but the treatment response is highly variable among different patients. The genetic underpinnings of the treatment response are not well understood. We performed a pharmacogenomic study of the treatment response to risperidone in patients with schizophrenia by using a SNP microarray -based genome-wide association study (GWAS) and whole exome sequencing (WES)-based GWAS. DNA samples were collected from 189 patients for the GWAS and from 222 patients for the WES after quality control in multiple centers of China. Antipsychotic response phenotypes of patients who received eight weeks of risperidone treatment were quantified with percentage change on the Positive and Negative Syndrome Scale (PANSS). The GWAS revealed a significant association between several SNPs and treatment response, such as three GRM7 SNPs (rs141134664, rs57521140, and rs73809055). Gene-based analysis in WES revealed 13 genes that were associated with antipsychotic response, such as GPR12 and MAP2K3. We did not identify shared loci or genes between GWAS and WES, but association signals tended to cluster into the GPCR gene family and GPCR signaling pathway, which may play an important role in the treatment response etiology. This study may provide a research paradigm for pharmacogenomic research, and these data provide a promising illustration of our potential to identify genetic variants underlying antipsychotic responses and may ultimately facilitate precision medicine in schizophrenia.

Supervised neural network models have achieved outstanding performance in the document summarization task in recent years. However, it is hard to get enough labeled training data with a high quality for these models to generate different types of summaries in reality. In this work, we mainly focus on improving the performance of the popular unsupervised Textrank algorithm that requires no labeled training data for extractive summarization. We first modify the original edge weight of Textrank to take the relative position of sentences into account, and then combine the output of the improved Textrank with K-means clustering to improve the diversity of generated summaries. To further improve the performance of our model, we innovatively incorporate external knowledge from open-source knowledge graphs into our model by entity linking. We use the knowledge graph sentence embedding and the tf-idf embedding as the input of our improved Textrank, and get the final score for each sentence by linear combination. Evaluations on the New York Times data set show the effectiveness of our knowledge-enhanced approach. The proposed model outperforms other popular unsupervised models significantly.