Microsoft Ignite kicks off soon and I thought it would be interesting to see what’s causing a buzz on Twitter by tracking some key hashtags related to the event.
I’m mainly interested in AI and Cloud technologies at the moment so want to track:
I’m using a tool I’ve built that integrates with social media APIs which can extract tweets and associated data that includes, but is not limited to:
- Tweet copy
- Tweet location
- Most mentioned user
- Most popular hashtag
- Most popular location
Here you can see the configuration being setup to track the hashtags:
This tool has integrations with the Azure Cognitive Services Text Analytics API which lets it perform:
- Named Entity Recognition – to detect the existence of people, places, organizations in each tweet
- Key Phrase Extraction – to return a list of strings and key talking points in each tweet
- Sentiment Analysis – to identify the underlying tone of each tweet
Historically, I had a custom Bayesian Classification API I had built but swapped this out in favour for the Text Analytics API for various reasons such as maintainability, accuracy, scalability and so on.
The data is visualised in a ASP.NET web application and in this screen shot, you can see some of the data that’s already being collected:
This screen shot shows you the underlying sentiment from the moment I tracked these hashtags:
Green is positive, red is negative – good vibes all round!
You can see it’s processed about 6,000 tweets in October and has processed about 1,000 tweets so far for November.
All of this is running locally on a spare laptop I have, crunching away 24×7. In the coming days I’ll be migrating it to Azure with an Azure SQL Database.
It’ll be interested to see the output after the end of the event and will share results with a follow up blog post!