Python for Data Science - Naive Bayes Classifiers

Chapter 6 - Other Popular Machine Learning Methods

Segment 5 - Naive Bayes Classifiers

Naive Bayes Classifiers

Naive Bayes is a machine learning method you can use to predict the likelihood that an event will occur given evidence that's present in your data.

Conditional Probability

\[P(B|A) = \frac{P(A and B)}{P(A)} \]

Tree Types of Naive Bayes Model

• Multinomial
• Bernoulli
• Gaussian

Naive Bayes Use Cases

• Spam Detection
• Customer Classification
• Credit Risk Protection
• Health Risk Protection

Naive Bayes Assumptions

Predictors are independent of each other.

A proiri assumption: the assumption the past conditions still hold true; when we make predictions from historical values we will get incorrect results if present circumstances have changed.

• All regression models maintain a priori assumption as well

import numpy as np
import pandas as pd
import urllib
import sklearn

from sklearn.model_selection import train_test_split
from sklearn import metrics
from sklearn.metrics import accuracy_score

from sklearn.naive_bayes import BernoulliNB
from sklearn.naive_bayes import GaussianNB
from sklearn.naive_bayes import MultinomialNB

Naive Bayes

Using Naive Bayes to predict spam

url = "https://archive.ics.uci.edu/ml/machine-learning-databases/spambase/spambase.data"

import urllib.request

raw_data = urllib.request.urlopen(url)
dataset = np.loadtxt(raw_data, delimiter=',')
print(dataset[0])

[  0.      0.64    0.64    0.      0.32    0.      0.      0.      0.
   0.      0.      0.64    0.      0.      0.      0.32    0.      1.29
   1.93    0.      0.96    0.      0.      0.      0.      0.      0.
   0.      0.      0.      0.      0.      0.      0.      0.      0.
   0.      0.      0.      0.      0.      0.      0.      0.      0.
   0.      0.      0.      0.      0.      0.      0.778   0.      0.
   3.756  61.    278.      1.   ]

X = dataset[:,0:48]

y = dataset[:,-1]

X_train, X_test, y_train, y_test = train_test_split(X,y,test_size=.2, random_state=17)

BernNB = BernoulliNB(binarize=True)
BernNB.fit(X_train, y_train)
print(BernNB)

y_expect = y_test
y_pred = BernNB.predict(X_test)

print(accuracy_score(y_expect, y_pred))

BernoulliNB(binarize=True)
0.8577633007600435

MultiNB = MultinomialNB()
MultiNB.fit(X_train, y_train)
print(MultiNB)

y_pred = MultiNB.predict(X_test)

print(accuracy_score(y_expect, y_pred))

MultinomialNB()
0.8816503800217155

GausNB = GaussianNB()
GausNB.fit(X_train, y_train)
print(GausNB)

y_pred = GausNB.predict(X_test)

print(accuracy_score(y_expect, y_pred))

GaussianNB()
0.8197611292073833

BernNB = BernoulliNB(binarize=0.1)
BernNB.fit(X_train, y_train)
print(BernNB)

y_expect = y_test
y_pred = BernNB.predict(X_test)

print(accuracy_score(y_expect, y_pred))

BernoulliNB(binarize=0.1)
0.9109663409337676