We are pleased to partner with NCF to bring the community an offer for 3 months of free access to Akkio’s no-code machine learning platform. With Akkio you can use your data to create and deploy machine learning models to predict the future of your business - without writing any code. With Akkio + Zapier, you can automate actions to take with your predictions.
Follow these instructions to set up your first AI-powered zap. This will lead you through the steps of connecting Twitter, Akkio, Slack, and Zapier to (1) find tweets that mention you, your company, or a keyword you specify, then (2) make a prediction in Akkio on the sentiment (positive or negative) of that tweet, then (3) send a notification of a negative tweet to Slack.
Once you have this working, it’s time to get creative and see what else you can predict, automate, and create!
To get going:
Set up a free Akkio account.
Clone this Zap into your Zapier account.
Now personalize the Zap to your Twitter, Akkio, and Zapier accounts.
Connect your Twitter account, select the trigger event, and the trigger search term for your zap (in this case, the search term could be your Twitter handle, your company name, etc).
One more step here - we need to filter out retweets so the Zap only fires when a new mention is made. So we need a filter with the following settings: Retweeted Status User Screen Name - Does Not Exist (as shown below).
Now let’s set up Akkio:
Click connect a new account
Browse over to your API keys on the team page by following the link, paste the keys into Zapier, then click “Yes, continue”.
Choose the Sentiment Model as the flow, and select the Twitter text option.
Then go ahead and test & continue:
Next, let’s set up our filter step. Choose “Make Prediction In Akkio” under “Only Continue if...” Then select “show all options”. Select “Probability Sentiment is Negative” and set the threshold to greater than 80% (so we only capture the really negative tweets and not the neutral ones - you can adjust this threshold if you like).
And your final confirmation screen should look like this:
Finally, let’s finish setting up the Zap by notifying a Slack channel when a negative tweet is detected. You can send the notifications to any slack channel - we just made a “twitter” channel. Add in the user name and the text of the tweet as the message text. The only other changes are to add an @channel notification (so the team gets alerted to act quickly) and drop a space between the user name and the text for legibility.
That’s it - run the test and turn on the Zap - you are now up and running with your own live Twitter Sentiment model.
To activate your 3 free months simply email the link to your Zap to “NCFPromo@akk.io” (please make sure to use the email you used when creating your Akkio Account) and we will turn on your free 3 months.
Then, get creative. Build a few different sentiment models - or use your data to create a machine learning model and automation custom to you. Make sure to tell us about your unique Akkio powered Zap for a chance to win a full 1-year subscription. The 3 winners will be selected by our judging panel - Josh from NCF, Andrew from Luhhu, and Jon from Akkio.