Python数据分析之微信好友数据分析

基于微信开放的个人号接口python库itchat，实现对微信好友的获取，并对省份、性别、微信签名做数据分析。

效果：

直接上代码，建三个空文本文件stopwords.txt，newdit.txt、unionWords.txt，下载字体simhei.ttf或删除字体要求的代码，就可以直接运行。

 #wxfriends.py  2018-07-09
import itchat
import sys
import pandas as pd
import matplotlib.pyplot as plt
plt.rcParams['font.sans-serif']=['SimHei']#绘图时可以显示中文
plt.rcParams['axes.unicode_minus']=False#绘图时可以显示中文
import jieba
import jieba.posseg as pseg
from scipy.misc import imread
from wordcloud import WordCloud
from os import path
#解决编码问题
non_bmp_map = dict.fromkeys(range(0x10000, sys.maxunicode + 1), 0xfffd)


#获取好友信息
def getFriends():
    friends = itchat.get_friends(update=True)[0:]
    flists = []
    for i in friends:
        fdict={}
        fdict['NickName']=i['NickName'].translate(non_bmp_map)
        if i['Sex'] == 1:
            fdict['Sex']='男'
        elif i['Sex'] == 2:
            fdict['Sex']='女'
        else:
            fdict['Sex']='雌雄同体'
        if i['Province'] == '':
            fdict['Province'] ='未知'
        else:
            fdict['Province']=i['Province']
        fdict['City']=i['City']
        fdict['Signature']=i['Signature']
        flists.append(fdict)
    return flists


#将好友信息保存成CSV
def saveCSV(lists):
    df = pd.DataFrame(lists)
    try:
        df.to_csv("wxfriends.csv",index = True,encoding='gb18030')
    except Exception as ret:
        print(ret)
    return df


#统计性别、省份字段    
def anysys(df):
    df_sex = pd.DataFrame(df['Sex'].value_counts())
    df_province = pd.DataFrame(df['Province'].value_counts()[:15])
    df_signature = pd.DataFrame(df['Signature'])
    return df_sex,df_province,df_signature


#绘制柱状图，并保存   
def draw_chart(df_list,x_feature):
    try:
        x = list(df_list.index)
        ylist = df_list.values
        y = []
        for i in ylist :
            for j in i:
                y.append(j)
        plt.bar(x,y,label=x_feature)
        plt.legend()
        plt.savefig(x_feature)
        plt.close()
    except:
        print("绘图失败")


#解析取个性签名构成列表     
def getSignList(signature):
    sig_list = []
    for i in signature.values:
        for j in i:
            sig_list.append(j.translate(non_bmp_map))
    return sig_list


#分词处理，并根据需要填写停用词、自定义词、合并词替换
def segmentWords(txtlist):
    stop_words = set(line.strip() for line in open('stopwords.txt', encoding='utf-8'))
    newslist = []
    #新增自定义词
    jieba.load_userdict("newdit.txt")
    for subject in txtlist:
        if subject.isspace():
            continue
        word_list = pseg.cut(subject)
        
        for word, flag in word_list:
            if not word in stop_words and flag == 'n' or flag == 'eng' and word !='span' and word !='class':
                newslist.append(word)
     #合并指定的相似词
    for line in open('unionWords.txt', encoding='utf-8'):
        newline = line.encode('utf-8').decode('utf-8-sig')    #解决\ufeff问题
        unionlist = newline.split("*")
        for j in range(1,len(unionlist)):
            #wordDict[unionlist[0]] += wordDict.pop(unionlist[j],0)
            for index,value in enumerate(newslist):
                if value == unionlist[j]:
                    newslist[index] = unionlist[0]  
    return newslist


#高频词统计
def countWords(newslist):
    wordDict = {}
    for item in newslist:
        wordDict[item] = wordDict.get(item,0) + 1
    itemList = list(wordDict.items())
    itemList.sort(key=lambda x:x[1],reverse=True)        
    for i in range(100):
        word, count = itemList[i]
        print("{}:{}".format(word,count))


#绘制词云
def drawPlant(newslist):
    d = path.dirname(__file__)
    mask_image = imread(path.join(d, "timg.png"))
    content = ' '.join(newslist)
    wordcloud = WordCloud(font_path='simhei.ttf', background_color="white",width=1300,height=620, max_words=200).generate(content)   #mask=mask_image,
    # Display the generated image:
    plt.imshow(wordcloud)
    plt.axis("off")
    wordcloud.to_file('wordcloud.jpg')
    plt.show()


def main():
    #登陆微信
    itchat.auto_login()   # 登陆后不需要扫码   hotReload=True
    flists = getFriends()
    fdf = saveCSV(flists)
    df_sex,df_province,df_signature = anysys(fdf)
    draw_chart(df_sex,"性别")
    draw_chart(df_province,"省份")
    wordList = segmentWords(getSignList(df_signature))
    countWords(wordList)
    drawPlant(wordList)
    
main()