opennlp中文训练模型分享
DoccatModel doccatModel = new DoccatModel(new FileInputStream("cn-name.bin"));
DocumentCategorizerME documentCategorizerME = new DocumentCategorizerME(doccatModel);
double[] categorize = documentCategorizerME.categorize(new String[]{"郑","明","海"});
测试结果:测试数据量11562 正确数9350 正确比例0.8086836187510811
数据来源 https://github.com/z17176/Chinese_conversation_sentiment