Chapter 4 of machine learning [series] naive Bayesian model

machine learning 【 series 】 Chapter 4 naive Bayesian model

Chapter four Naive bayesian model

---

---

Preface

This chapter mainly explains the naive Bayesian model in machine learning , First, explain the algorithm principle and programming implementation of naive Bayes , Then through a classic case ----- Tumor prediction model to consolidate the knowledge .

---

One 、 Algorithm principle of naive Bayesian model

• 1. Bayesian model with one-dimensional characteristic variables

   P(Y|X1) = P(X1|Y)P(Y)/P(X1)
    In Bayesian formula P(X1) It's called a priori probability ,P(Y|X1) It's called posterior probability , and P(X1|Y) It's called likelihood .
  

• 2. Bayesian model under two-dimensional characteristic variables

   P(Y|X1,X2) = P(X1,X2|Y)P(Y)/P(X1,X2)
  

• n Bayesian model under dimensional characteristic variables

    P(Y|X1,X2,.....,Xn) = P(X1,X2,....,Xn)P(Y)/P(X1,X2,......,Xn)
     What this formula means is ： In character X1, features X2, features X3 And so on , Category Y Probability of occurrence .
  

Two 、 Simple code implementation of naive Bayesian model

The simple code implementation of Bayesian model is introduced through the following code （ The Gaussian Bayesian classifier is used here ）

from sklearn.naive_bayes import GaussianNB
X = [[1,2],[3,4],[5,6],[7,8],[9,10]]
y = [0,0,0,1,1]

model = GaussianNB()
model.fit(X,y)
print(model.predict([[5,5]]))

---> The output is :
[0]

case analysis ： Tumor prediction model

1. Reading data

import pandas as pd
df = pd.read_excel(" Tumor data .xlsx")

2. Divide characteristic variables and target variables

X = df.drop(columns=" The nature of the tumor ")
y = df[" The nature of the tumor "]

3. Model construction and use

#  Divide the training set and the test set 
from sklearn.model_selection import train_test_split
# X_train,y_train Data of characteristic variables and target variables in the training set ,X_test,y_test Is the characteristic variable and target variable data in the test set 
# test_size=0.2, Press 8:2 To divide the training set and the test set 
X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.2,random_state=1)  # train_test_split() Functions are randomly partitioned 

#  Model structures, 
from sklearn.naive_bayes import GaussianNB
nb_clf = GaussianNB()   #  Gaussian naive Bayesian model 
# fit() Function training model , The parameters passed in are the previously obtained training set data X_train,y_train
nb_clf.fit(X_train,y_train)

# 3. Model prediction and evaluation 
# nb_clf It is the naive Bayesian model built above 
y_pred = nb_clf.predict(X_test)
#  You can view the front... Of the forecast results through printout 100 term 
print(y_pred[:100])

#  utilize DataFrame Relevant knowledge points of , Summarize the forecast results y_pred And the actual values in the test set y_test
a = pd.DataFrame()
a[" Predictive value "] = list(y_pred)
a[" actual value "] = list(y_test)
print(a.head())

#  View the prediction accuracy of all test set data 
from sklearn.metrics import accuracy_score
score = accuracy_score(y_pred,y_test)
print(score)

---

summary

Reference books ：《Python Big data analysis and machine learning business case practice 》