나이브베이즈 분류 #나이브 베이즈#확률통계

1. 나이브 베이즈 확률

def main():
    sensitivity = float(input())
    prior_prob = float(input())
    false_alarm = float(input())

    print("%.2lf%%" % (100 * mammogram_test(sensitivity, prior_prob, false_alarm)))

def mammogram_test(sensitivity, prior_prob, false_alarm):
    p_a1_b1 = sensitivity # p(A = 1 | B = 1)

    p_b1 = prior_prob    # p(B = 1)

    p_b0 = 1-prior_prob    # p(B = 0)

    p_a1_b0 = false_alarm # p(A = 1|B = 0)

    p_a1 = p_b0*p_a1_b0+p_b1*p_a1_b1    # p(A = 1)

    p_b1_a1 = p_a1_b1*p_b1/p_a1 # p(B = 1|A = 1)

    return p_b1_a1

if __name__ == "__main__":
    main()

 

2. 나이브베이즈 분류기

# [나이브베이즈 분류기]

import re
import math
import numpy as np

def main():
    M1 = {'r': 0.7, 'g': 0.2, 'b': 0.1} # M1 기계의 사탕 비율
    M2 = {'r': 0.3, 'g': 0.4, 'b': 0.3} # M2 기계의 사탕 비율
    
    test = {'r': 4, 'g': 3, 'b': 3}

    print(naive_bayes(M1, M2, test, 0.7, 0.3))

def naive_bayes(M1, M2, test, M1_prior, M2_prior):
    a=M1_prior*((M1["r"]**test["r"])*(M1["g"]**test["g"])*(M1["b"]**test["b"]))
    b=M2_prior*((M2["r"]**test["r"])*(M2["g"]**test["g"])*(M2["b"]**test["b"]))
    mySum=a+b

    return [a/mySum, b/mySum]

if __name__ == "__main__":
    main()

 

3. 감정분석기 bag of words

# [감정분석기 bag of Words]

import re

special_chars_remover = re.compile("[^\w'|_]")

def main():
    sentence = input()
    bow = create_BOW(sentence)

    print(bow)


def create_BOW(sentence):
    sentence=sentence.lower()
    sentence=remove_special_characters(sentence)
    sentence=sentence.split()
    
    bag={}
    for word in sentence:
        if len(word)<1:
            continue
            
        if word in bag:
            bag[word]+=1
        else:
            bag[word]=1
    
    return bag
    

def remove_special_characters(sentence):
    return special_chars_remover.sub(' ', sentence)


if __name__ == "__main__":
    main()