从音调分析器的JSON响应中从字典列表中提取数据

Question

我正在使用IBM的声调分析器分析文本，并试图提取与句子音调有关的所有信息(例如：sentence_id、text、tones、tone_id、tone_name、score)，并将其添加到数据帧中(有列、sentence_id、text、tones、tone_id、score和tone_name)。这是我的输出的一个示例：

​

> [{'document_tone': {'tones': [{'score': 0.551743,
     'tone_id': 'analytical',
     'tone_name': 'Analytical'}]},
  'sentences_tone': [{'sentence_id': 0,
    'text': '@jozee25 race is the basis on which quotas are implemented.',
    'tones': []},
   {'sentence_id': 1, 'text': 'helloooooo', 'tones': []}]},
 {'document_tone': {'tones': []}},
 {'document_tone': {'tones': [{'score': 0.802429,
     'tone_id': 'analytical',
     'tone_name': 'Analytical'},
    {'score': 0.60167, 'tone_id': 'confident', 'tone_name': 'Confident'}]},
  'sentences_tone': [{'sentence_id': 0,
    'text': '@growawaysa @cricketandre i have the answer on top yard from dpw:it is not currently "surplus to govt requirements".it is still being used for garaging until a new facility is ready in maitland.the',
    'tones': [{'score': 0.631014,
      'tone_id': 'analytical',
      'tone_name': 'Analytical'}]},
   {'sentence_id': 1,
    'text': 'cost of the housing options will of course depend on prospects for cross subsidisation.',
    'tones': [{'score': 0.589295,
      'tone_id': 'analytical',
      'tone_name': 'Analytical'},
     {'score': 0.509368, 'tone_id': 'confident', 'tone_name': 'Confident'}]}]},
 {'document_tone': {'tones': [{'score': 0.58393,
     'tone_id': 'tentative',
     'tone_name': 'Tentative'},
    {'score': 0.641954, 'tone_id': 'analytical', 'tone_name': 'Analytical'}]}},
 {'document_tone': {'tones': [{'score': 0.817073,
     'tone_id': 'joy',
     'tone_name': 'Joy'},
    {'score': 0.920556, 'tone_id': 'analytical', 'tone_name': 'Analytical'},
    {'score': 0.808202, 'tone_id': 'tentative', 'tone_name': 'Tentative'}]},
  'sentences_tone': [{'sentence_id': 0,
    'text': 'thanks @khayadlangaand colleagues for the fascinating tour yesterday.really',
    'tones': [{'score': 0.771305, 'tone_id': 'joy', 'tone_name': 'Joy'},
     {'score': 0.724236, 'tone_id': 'analytical', 'tone_name': 'Analytical'}]},
   {'sentence_id': 1,
    'text': 'eyeopening and i learnt a lot.',
    'tones': [{'score': 0.572756, 'tone_id': 'joy', 'tone_name': 'Joy'},
     {'score': 0.842108, 'tone_id': 'analytical', 'tone_name': 'Analytical'},
     {'score': 0.75152, 'tone_id': 'tentative', 'tone_name': 'Tentative'}]}]},

​

这是我为获得这个输出而编写的代码：

result =[]
for i in helen['Tweets']:
   tone_analysis = tone_analyzer.tone(
       {'text': i},
       'application/json'
   ).get_result()
   result.append(tone_analysis)

Answer 1

首先，由于您的JSON格式不太好，所以我使用的是来自音调分析器API引用的这里。

使用API引用和Pandas json_normalize中的JSON，下面是我提出的代码

from pandas.io.json import json_normalize

jsonfile = {
  "document_tone": {
    "tones": [
      {
        "score": 0.6165,
        "tone_id": "sadness",
        "tone_name": "Sadness"
      },
      {
        "score": 0.829888,
        "tone_id": "analytical",
        "tone_name": "Analytical"
      }
    ]
  },
  "sentences_tone": [
    {
      "sentence_id": 0,
      "text": "Team, I know that times are tough!",
      "tones": [
        {
          "score": 0.801827,
          "tone_id": "analytical",
          "tone_name": "Analytical"
        }
      ]
    },
    {
      "sentence_id": 1,
      "text": "Product sales have been disappointing for the past three quarters.",
      "tones": [
        {
          "score": 0.771241,
          "tone_id": "sadness",
          "tone_name": "Sadness"
        },
        {
          "score": 0.687768,
          "tone_id": "analytical",
          "tone_name": "Analytical"
        }
      ]
    },
    {
      "sentence_id": 2,
      "text": "We have a competitive product, but we need to do a better job of selling it!",
      "tones": [
        {
          "score": 0.506763,
          "tone_id": "analytical",
          "tone_name": "Analytical"
        }
      ]
    }
  ]
}

mydata = json_normalize(jsonfile['sentences_tone'])
mydata.head(3)
print(mydata)

tones_data = json_normalize(data=jsonfile['sentences_tone'], record_path='tones')
tones_data.head(3)
print(tones_data)

输出数据将是

   sentence_id                        ...                                            tones
0            0                        ...[{'score': 0.801827, 'tone_id': 'analytical', ...
1            1                        ...[{'score': 0.771241, 'tone_id': 'sadness', 'to...
2            2                        ...[{'score': 0.506763, 'tone_id': 'analytical', ...

[3 rows x 3 columns]
      score     tone_id   tone_name
0  0.801827  analytical  Analytical
1  0.771241     sadness     Sadness
2  0.687768  analytical  Analytical
3  0.506763  analytical  Analytical

此外，我还为您创建了REPL，以更改输入并在browser - https://repl.it/@aficionado/DarkturquoiseUnnaturalDistributeddatabase上运行代码。

请参阅此Kaggle链接以了解有关使用Pandas在Python中扁平JSON的更多信息