Are Tweets exactly the same as the ones Twitter Sends when Streaming? For instance do they have the same data and structure, especially regarding ReTweets and Mentions.
Some fields are named slightly differently, but we do pass on the vast majority of the data we receive from Twitter, plus additional augmentations, such as fully resolved links (including link metadata), Salience information, and more.
You can see some sample Tweets we would pass along on our Output Data page, or by creating and running your own stream, and analysing the results.
Feel free to get in touch if you would like to request more information or a demo.
Thanks for the reply. I’m after the
- The data in entities -> user_mentions, datasift passes a normalised version
- What happens to 'retweeted_status', does this get converted to something else? I need to know who was the original Tweeter and the Re-Tweeters tweet.
1. User Mentions
We have a blog post about How Best to Filter for Twitter @Mentions. Twitter @mentions are returned as both mentioned @usernames and user IDs as their own entities
Here is an example retweeted status. Data about the original Tweeter is contained in the twitter.retweeted.* fields, and the user creating the retweet, and the retweet data, are contained within the twitter.retweet.* fields.