NER for Chinese electronic medical records. Use doc2vec, self_attention and multi_attention.
Medical Named Entity Recognition implement using bi-directional lstm and crf model with char embedding.CCKS2017中文电子病例命名实体识别项目,主要实现使用了基于字向量的四层双向LSTM与CRF模型的网络.该项目提供了原始训练数据样本(一般醒目,出院情况,病史情况,病史特点,诊疗经过)与转换版本,训练脚本,预训练模型,可用于序列标注研究.把玩和PK使用.
Text classification using different neural networks (CNN, LSTM, Bi-LSTM, C-LSTM).
Depend on the GPS trajectory dataset, We do location recommendation and user recommendation.
Natural Language Processing with TensorFlow, published by Packt
This code implements a simple CNN model for document classification with tensorflow.
AI100竞赛:http://competition.ai100.com.cn/html/game_det.html?id = 24&tab = 1 的代码,主要用于文本分类,其中涉及CHI选择特征词,TFIDF计算权重,朴素贝叶斯,决策树,SVM,XGBoost等算法