python机器学习-课堂测试5的决策树分析

将商家的特征值提取出来，转成CSV文件

 

#问题企业决策树
def problem():
    # 1.获取数据
    data_titanic = pd.read_csv("key_data_fp.csv")
    # 2.获取目标值与特征值
    x = data_titanic[[ "xf_count", "gf_count","del_count","zfcs"]]
    y = data_titanic["problem"]
    # 3.数据处理
    # 1).缺失值处理
    # x["age"].fillna(x["age"].mean(), inplace=True)  # 填补处理dropna()删除缺失值所在的行
    # 2).转换为字典
    x = x.to_dict(orient="records")
    # 4.划分数据集
    x_train, x_test, y_train, y_test = train_test_split(x, y, random_state=22)
    # 5.字典特征抽取
    transfer = DictVectorizer()
    x_train = transfer.fit_transform(x_train)
    x_test = transfer.transform(x_test)
    # 6.决策树预估器
    estimator = DecisionTreeClassifier(criterion="entropy")  # criterion默认为gini系数，此处选择的为信息增益的熵
    # max_depth:树深的大小,random_state：随机数种子
    estimator.fit(x_train, y_train)
    # 7.模型评估
    y_predict = estimator.predict(x_test)
    print("直接对比真实值和预测值:\n", y_test == y_predict)
    score = estimator.score(x_test, y_test)
    print("准确率为：\n", score)
    # # 8.决策树可视化
    # export_graphviz(estimator, out_file="titanic_tree.dot", feature_names=transfer.get_feature_names())
    # # 使用随机森林
    # estimator = RandomForestClassifier()