MachineLearning.py 文件源码

def jaccard_similarity_score(context1, context2, flag1, flag2):
    #print 'context1', context1
    try:
        if flag1 and len(context1)!=0:
            temp = context1[-1]
            context1.pop()
            context1 += list(pseg.cut(temp))
        if flag2 and len(context1)!=0:
            temp = context2[-1]
            context2.pop()
            context2 += list(pseg.cut(temp))
    except:
        pass

    mySet = set(context1 + context2)
    a1 = []
    a2 = []
    for item in mySet:
        if item in context1:
            a1.append(1)
        else:
            a1.append(0)
        if item in context2:
            a2.append(1)
        else:
            a2.append(0)
    #print sklearn.metrics.jaccard_similarity_score(a1,a2)
    return sklearn.metrics.jaccard_similarity_score(a1,a2)

# element[i]?element[j]?contextSim