Article directory
Experimental content: Tumor classification and prediction (Naive Bayes)
The naive Bayes method is used to classify the breast cancer diagnosis data set in Wisconsin, USA, and implement a classifier for breast cancer detection to determine whether a patient's tumor is benign or malignant.
Experimental requirements
1. Import the data set that comes with sklearn: Wisconsin breast tumor data set (load_breast_cancer).
2. Print the data set keys (keys) to view the information contained in the data set.
3. Print and view the tumor classification (target_names) and tumor feature names (feature_names) marked in the data set.
4. Split the data set into a training set and a test set, and print and view the data shape of the training set and test set.
5. Configure the Gaussian Naive Bayes model.
6. Train the model.
7. Evaluate the model and print to view the model scores (print the scores of the training set and test set respectively).
8. Model prediction: Select a certain sample for prediction. (Can make multiple predictions for different samples)
[Dataset]: Wisconsin Breast Tumor Dataset
Tumor in this data set is a very classic data set used for medical condition analysis, including 569 case data samples, each sample has 30 features.
Samples are divided into two categories: malignant and benign.
The features of this dataset were computed from digitized fine-needle aspiration (FNA) images of a breast mass. They describe the characteristics of the cell nuclei in the image.
Complete code
#导入包
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.datasets import load_breast_cancer
from sklearn.model_selection import train_test_split
from sklearn.model_selection import cross_val_score
from sklearn.metrics import classification_report#导入分类报告模板
from sklearn.naive_bayes import GaussianNB#导入先验概率的高斯朴素贝叶斯模型
from sklearn.naive_bayes import MultinomialNB#导入先验概率为多项式分布的朴素贝叶斯模型
from sklearn.naive_bayes import BernoulliNB#导入先验概率为伯努利分布的朴素贝叶斯模型
#1.导入sklearn自带的数据集：威斯康星乳腺肿瘤数据集（load_breast_cancer）。
breast_cancer=load_breast_cancer()
#2.打印数据集键值（keys），查看数据集包含的信息。
print(breast_cancer.keys())
#3.打印查看数据集中标注好的肿瘤分类（target_names）、肿瘤特征名称（feature_names）。
print(breast_cancer.target_names)
print(breast_cancer.feature_names)
#4.将数据集拆分为训练集和测试集，打印查看训练集和测试集的数据形态（shape）。
data=pd.DataFrame(breast_cancer.data)
target=pd.DataFrame(breast_cancer.target)
X=np.array(data.values)
y=np.array(target.values)
train_X,test_X,train_y,test_y=train_test_split(X,y,test_size=0.2)
print(train_X.shape,test_X.shape)
#5.配置高斯朴素贝叶斯模型。
model=GaussianNB()
#6.训练模型。
model.fit(train_X,train_y)
#7.评估模型，打印查看模型评分（分别打印训练集和测试集的评分）。
#cross_val_score交叉验证
#计算高斯朴素贝叶斯算法模型的准确率
sorce=cross_val_score(model,train_X,train_y,cv=10,scoring='accuracy')
print("高斯朴素贝叶斯模型的准确率:",sorce.mean())
#打印训练集和测试集的评分
print("高斯朴素贝叶斯模型训练集的评分:",model.score(train_X,train_y))
print("高斯朴素贝叶斯模型测试集的评分:",model.score(test_X,test_y))
#8.模型预测：选取某一样本进行预测。（可以进行多次不同样本的预测）
pre_y=model.predict(test_X)
# sklearn中的classification_report函数用于显示主要分类指标的文本报告．在报告中显示每个类的精确度，召回率，F1值等信息。
print(classification_report(test_y,pre_y))