机器学习手记--

s

# -*- coding: utf-8 -*-
from numpy import *
from numpy import linalg as la


def loadData():           
       a=  [[0, 0, 0, 2, 2],
           [0, 0, 0, 3, 3],
           [0, 0, 0, 1, 1],
           [1, 1, 1, 0, 0],
           [2, 2, 2, 0, 0],
           [5, 5, 5, 0, 0],
           [1, 1, 1, 0, 0]]
       return mat(a)
       
def ecludianSim(vecA,vecB):
    return 1/(1+la.norm(vecA-vecB))
    
def cosSim(vecA,vecB):
    upper=vecA*vecB
    lower=float(la.norm(vecA)*la.norm(vecB))
    return (1+upper/lower)/2
    
    
    
def recommend(dataMat,userID):
    
    userRow=dataMat[userID,:]
    notRatedIndexes=nonzero(userRow.A==0)
    unRatedObjectScore={}
    for i in notRatedIndexes[1]:#i为未评分物品
        score=scoreCompute(userID, i,dataMat)
        unRatedObjectScore[i]=score
    return unRatedObjectScore

        
def scoreCompute(userID,i,dataMat):
    simTotal=0.0
    allTotal=0.0
    for c in range(dataMat.shape[1]):
        if (dataMat[userID,c]>0):
            overlap=nonzero(logical_and( dataMat[:,i],dataMat[:,c]))[0]  #logical_and  评判是否为0 
            if overlap==0:
                sim=0      
            sim=ecludianSim(dataMat[overlap,i],dataMat[overlap,c])
            simTotal+=sim
            allTotal+=sim*dataMat[userID,c]
    if  simTotal==0: #没有相似的，全都回府
        return 0
    finalScore=allTotal/simTotal
    return finalScore