[Python知识库]python数据预处理——关联数据写入csv文件，为画图作准备

python数据预处理——有关联数据写入csv文件，为画图作准备

目标格式如下：

测试代码：

from pandas import  DataFrame
# import json
# import numpy as np
#
# file = open('clean_data.txt', encoding='UTF-8')
# js = file.read()
# dic = json.loads(js)
# #print(dic)
# file.close()



authors=[]
authors_id={}
empty=[]
author_name={}

#第一个循环代表遍历每一篇论文
author_name['id']=123456
author_name['name']='xiaotang'
author_name['follow_id']=1234567
author_name['follow_name']='xiaotangtang'
authors.append(author_name)
author_name={}
author_name['id']=12345678
author_name['name']='xiaotang123'
author_name['follow_id']=123456789
author_name['follow_name']='xiaotangtang456'
authors.append(author_name)
print(authors)
data_frame = DataFrame(data=authors)
data_frame.to_csv("a.csv")

筛选数据集中的数据：

from pandas import  DataFrame
import json
import numpy as np

file = open('clean_data.txt', encoding='UTF-8')
js = file.read()
dic = json.loads(js)
#print(dic)
file.close()

count=0

authors=[]
empty=[]
author_name={}

#第一个循环代表遍历每一篇论文
for leng_dic in range(len(dic)-1):
    author_dic=dic[leng_dic].get('authors', [])
    if author_dic is None:
        empty.append(leng_dic)
    else:

        author_name_now = []
        author_id_now = []


        #遍历每篇论文里的作者信息
        for i in range( len(author_dic)-1):
            try:
                author_id_now.append(author_dic[i]['_id'])
            except:
                author_id_now.append([])


            try:
                author_name_now.append(author_dic[i]['name'])
            except:
                author_id_now.append([])


        for j in range(len(author_name_now)-1):
            for t in range(len(author_name_now) - 1):
                author_name={}

                if j !=t:
                    #print(123)
                    author_name['id']=author_id_now[j]
                    author_name['name']= author_name_now[j]
                    author_name['follow_id']=author_id_now[t]
                    author_name['follow_name'] = author_name_now[t]
                    authors.append(author_name)
                    #print('start')
                    #print(authors)


data_frame = DataFrame(data=authors)
data_frame.to_csv("tu1.csv")

部分数据集没有部分关键字，直接删掉：

from pandas import  DataFrame
import json
import numpy as np

file = open('clean_data.txt', encoding='UTF-8')
js = file.read()
dic = json.loads(js)
#print(dic)
file.close()

count=0

authors=[]
empty=[]
author_name={}

#第一个循环代表遍历每一篇论文
for leng_dic in range(len(dic)-1):
    author_dic=dic[leng_dic].get('authors', [])
    if author_dic is None:
        empty.append(leng_dic)
    else:

        author_name_now = []
        author_id_now = []


        #遍历每篇论文里的作者信息
        for i in range( len(author_dic)-1):
            try:
                author_id_now.append(author_dic[i]['_id'])
                author_name_now.append(author_dic[i]['name'])
            except:
                pass




        for j in range(len(author_name_now)-1):
            for t in range(len(author_name_now) - 1):
                author_name={}

                if j !=t:
                    #print(123)
                    author_name['id']=author_id_now[j]
                    author_name['name']= author_name_now[j]
                    author_name['follow_id']=author_id_now[t]
                    author_name['follow_name'] = author_name_now[t]
                    authors.append(author_name)
                    #print('start')
                    #print(authors)


data_frame = DataFrame(data=authors)
data_frame.to_csv("tu1.csv")