Chinese Q&A Community Medical Entity Recognition with Character-Level Features and Self-Attention Mechanism

Abstract

With the rapid development of Internet, the medical Q&A community has become an important channel for people to obtain and share medical and health knowledge. Online medical entity recognition (OMER), as the foundation of medical and health information extraction, has attracted extensive attention of researchers in recent years. In order to further improve the research progress of Chinese OMER, LSTM-Att-Med model is proposed in this paper to capture more external semantic features and important information. First, Word2vec is used to generate the character-level vectors with semantic features on the basis of the unlabeled corpus in the medical domain and open domain respectively. Then, the two character-level vectors are embedded into BiLSTM-CRF as features to construct LSTM-Wiki and LSTM-Med models. Finally, Self-Attention mechanism is introduced into LSTM-Med model, and the performance of the model is validated by using the self-labeled data. The 10-fold cross-validation experiment shows that LSTM-Att-Med with Self-Attention mechanism introduced achieves the best performance and the F-value can be up to 91.66%, which is 0.72% higher than that of BiLSTM-CRF. In addition, the experiment result demonstrates that the improvements of F-value are inconsistent for different corpora based on LSTM-Att-Med. The paper also analyzes the recognition performance and error results of different medical entities.

A Mixed Method for Feature Extraction Based on Resonance Filtering

“…Some scholars are devoted to the study of image feature extraction [1,2]. Feature extraction can be used for image classification [3,4], image segmentation [4][5][6][7], target detection [8][9][10][11], attention mechanism of the visual system [11][12][13][14][15][16] and other research directions. Images have individual features and common features, which are adversarial and interdependent in image recognition.…”

A Hybrid Approach for Network Rumor Detection Based on Attention Mechanism and Bidirectional GRU Model in Big Data Environment

“…Nearly 50,000 pieces of data are crawled in the first part, which include both rumor data and nonrumor data (half of each). In order to ensure the reliability of information, relevant nonrumor data are crawled from major channels, while rumor data are crawled mainly through messages released by official rumor-denial microblog accounts [20]. The second part is about the rumor dissemination data.…”