KNN 알고리즘 분류 영화 유형

'''

    KNN  ：
        1.                
        2.         (      )
        3.       K  
        4.  K             
        5.             

                                 ，            
        KNN    k             ，            

'''

    KNN  ：
        1.                
        2.         (      )
        3.       K  
        4.  K             
        5.             

                                 ，            
        KNN    k             ，            



'''
import numpy as np

#     
def creatDataSet():
    #                   
    group = np.array([[3,10,23,114],
                      [2,5,23,160],
                      [1,9,8,154],
                      [101,10,12,11],
                      [99,5,7,8,],
                      [98,2,10,5],
                      [4,99,14,10],
                      [6,100,11,23],
                      [9,100,1,1],
                      [1,4,99,3],
                      [4,5,78,6],
                      [1,4,88,9]])
    labels = ["   ","   ","   ","   ","   ","   ","   ","   ","   ","   ","   ","   "]
    return group, labels

def classify(input, dataSet, labels, k):
    #        
    dataSize = dataSet.shape[0]
    print(dataSet.shape[0])
    #             
    # nump.tile()             
    print(input)
    kdiff = np.tile(input, (dataSize, 1)) - dataSet
    sqdiff = kdiff**2
    #        ，           
    sqnumdiff = np.sum(sqdiff, axis=1)
    dist = sqnumdiff**0.5
    #        
    soreDistIndex = np.argsort(dist)
    classCount = {}
    for i in range(k):
        print(soreDistIndex[i])
        voteLabel = labels[soreDistIndex[i]]
        print(voteLabel)
        #     K              
        classCount[voteLabel] = classCount.get(voteLabel, 0) + 1
    #               
    maxCount = 0
    for key, value in classCount.items():
        if value > maxCount:
            maxCount = value
            classes = key

    return classes

if __name__ == "__main__":
    dataSet, labels = creatDataSet()
    #     
    input = np.array([9,90,500,4])
    k = 3
    output = classify(input, dataSet, labels, k)
    print("     :", input, "     ：", output)