Sentiment Analysis

rpi.analyticsdojo.com

47. Sentiment Analysis

• The pricing.

• Helpful example.

• Microsoft quickstart documentation

#This imports some required packages.
#import utils #Often I'll develop functions in a notebook then move to utils. 
import pandas as pd
import urllib.request
import json
import pprint

48. Keys are Needed for the API.

You will find the appropriate key on the slack channel.

#I typically store my config values. 
azure_text_endpoint= 'https://eastus.api.cognitive.microsoft.com/text/analytics/v2.0'
azure_text_key1= 'key posted on slack'

def azure_request(command, endpoint, key, postdata):
    #Set URI
    uri=endpoint+"/"+command
    #Set header
    headers = {}
    headers['Ocp-Apim-Subscription-Key'] = key
    headers['Content-Type'] = 'application/json'
    headers['Accept'] = 'application/json'
    #Make request
    request = urllib.request.Request(uri, postdata, headers)
    response = urllib.request.urlopen(request)
    return json.loads(response.read().decode('utf-8'))

def df_to_postdata(df):
    return json.dumps({'documents': json.loads(df.to_json(orient='records')) }).encode('utf-8')

#First lets test with sample data from the examples. 
#https://docs.microsoft.com/en-us/azure/cognitive-services/text-analytics/quickstarts/python
test_data = json.dumps({ 'documents': [
    { 'id': '1', 'language': 'en', 'text': 'I really enjoy the new XBox One S. It has a clean look, it has 4K/HDR resolution and it is affordable.' },
    { 'id': '2', 'language': 'es', 'text': 'Este ha sido un dia terrible, llegué tarde al trabajo debido a un accidente automobilistico.' }
]}).encode('utf-8')
pprint.pprint(test_data)

(b'{"documents": [{"id": "1", "language": "en", "text": "I really enjoy the new'
 b' XBox One S. It has a clean look, it has 4K/HDR resolution and it is afforda'
 b'ble."}, {"id": "2", "language": "es", "text": "Este ha sido un dia terrible,'
 b' llegu\\u00e9 tarde al trabajo debido a un accidente automobilistico."}]}')

test_result=azure_request('sentiment', azure_text_endpoint, azure_text_key1, test_data)
test_result

{'documents': [{'id': '1', 'score': 0.9619677066802979},
  {'id': '2', 'score': 0.024017512798309326}],
 'errors': []}