1. sklearn Chinese text classification(Naive Bayes, Logistic Regression, XGboost, lightGBM, SVM, KNN,
Random Forest, Decision Tree, GBDT)to test the result of bio-text classification,the data i use(neg.txt, pos.txt)
Naive Bayes | 0.9032786885245901 |
---|---|
Logistic Regression | 0.8737704918032787 |
SVM | 0.8836065573770492 |
Xgboost | 0.8442622950819673 |
LightGBM | 0.8508196721311475 |
knn | 0.6081967213114754 |
RFC | 0.8721311475409836 |
DT | 0.8 |
GBDT | 0.8508196721311475 |