Correction: Lesson 5 Lab 2: Translation and Sentiment Analysis

Text translation


Question: Several tweets are not in English. How can you easily detect the language?

Answer: look at the user_lang field

 

Question: What is the URL you should build to query the server?

Answer: http://api.mymemory.translated.net/get?q=<text>&langpair=en|it

 

Question: How do you convert this query in GREL?

Answer: based on the text column
"http://api.mymemory.translated.net/get?q="+escape(value, 'url')+"&langpair="+cells['user_lang'].value+"|en"

 

Select 15 tweets in non-English and process them through the service.

 

Question: Extract the translation and quality score:

Anwer:

  • score: value.parseJson().responseData.match

  • translation: value.parseJson().responseData.translatedText
Last modified: Saturday, 6 April 2019, 12:44 PM