Medical Text Classification System Based on Deep Learning

With the continuous development of the Internet, text information as the most common form of expression has naturally become a research hotspot, among which medical records and clinical diagnosis are important medical resources. However, due to data sparseness and high latitude, medical text classification is more challenging and difficult than other fields. This paper proposes a medical text classification model that combines an improved bidirectional long-term short-term memory (BI-LSTM) and attention mechanism. BI-LSTM is used to extract features and contains sentence information, and the attention mechanism is used to obtain different BI-LSTMs to encode sentence weights, and the attention mechanism is used to decode them. Finally, we use the SoftMax classifier to obtain the classification results. This article uses five public data sets and self-built medical data sets for experimental comparison and verification to verify the effectiveness of the method.