导入JSON文件进行Python分析

Question

我想在Python编辑器中导入一个JSON文件，以便对数据进行分析。我对Python还很陌生，所以不太清楚该怎么做。我的JSON文件里全是推文数据，示例在这里：

{"id":441999105775382528,"score":0.0,"text":"blablabla","user_id":1441694053,"created":"Fri Mar 07 18:09:33 GMT 2014","retweet_id":0,"source":"<a href=\"http://twitterfeed.com\" rel=\"nofollow\">twitterfeed</a>","geo_long":null,"geo_lat":null,"location":"","screen_name":"SevenPS4","name":"Playstation News","lang":"en","timezone":"Amsterdam","user_created":"2013-05-19","followers":463,"hashtags":"","mentions":"","following":1062,"urls":"http://bit.ly/1lcbBW6","media_urls":"","favourites_count":4514,"reply_status_id":0,"reply_user_id":0,"is_truncated":false,"is_retweet":false,"original_text":null,"status_count":4514,"description":"Tweeting the latest Playstation news!","url":null,"utc_offset":3600}

我有几个问题：

我该如何导入这个JSON文件，以便在Python编辑器中对它进行分析呢？

我怎么才能只分析一部分数据（比如100条或200条，而不是全部）呢？

有没有办法在不手动处理所有数据的情况下，去掉一些字段，比如score、user_id、created等？

有些推文里面有无效或无法使用的符号，有没有办法在不手动处理的情况下去掉这些符号呢？

Answer 1

我会推荐使用Pandas来完成这个工作，因为你不仅需要加载json文件，还要对数据进行一些分析。根据你的json文件的大小，这段代码应该可以满足你的需求：

import pandas as pd
import json

# read a sample json-file (replace the link with your file location
j = json.loads("yourfilename")
# you might select the relevant keys before constructing the data-frame
df = pd.DataFrame.from_dict([{k:v} for k,v in j.iteritems() if k in ["id","retweet_count"]])
# select a subset (the first five rows)
df.iloc[:5]
# do some analysis
df.retweet_count.sum()
>>> 200