特征选取之IV（信息值）及python实现

IV表征特征的预测能力：小于0.02，几乎没有预测能力；小于0.1，弱；小于0.3，中等；小于0.5，强；大于0.5，难以置信，需进一步确认

WOE describes the relationship between a predictive variable and a binary target variable.
IV measures the strength of that relationship.

计算公式：暂不写……

代码实现如下：

# 定义字典，记录每个特征的信息值iv
iv_dict=dict()
def cal_iv(df,feature,target='target'):
    '''
    用于二分类的信息值计算，返回信息值和具体信息
    :df pd.DataFrame
    :feature 选择的特征
    :target 目标特征名
    '''
    ls=[]
    for val in df[feature].unique():
        al=df[df[feature]==val][feature].count()
        good=df[(df[feature]==val)&(df[target]==1)][feature].count()
        bad=df[(df[feature]==val)&(df[target]==0)][feature].count()
        ls.append([val,al,good,bad])
    data=pd.DataFrame(ls,columns=[feature,'all','good','bad'])
    good_rate=data['good']/data['good'].sum()# good边际概率
    bad_rate=data['bad']/data['bad'].sum()# bad边际概率
    data['woe']=np.log(good_rate/bad_rate)# woe为证据权重
    data = data.replace({'woe': {np.inf: 0, -np.inf: 0}})
    data['iv']=data['woe']*(good_rate-bad_rate)
    iv=data.iv.sum()
#     添加到字典
    if feature not in iv_dict.keys():
        iv_dict[feature]=iv
    print('iv for %s is %f: '%(feature,iv))
    return iv,data