Understanding how ‘we’ feel today: Automated Social Sentiment

I wrote last week about TapDinGo, an online travel concierge that uses social media sentiment to go beyond curated content to build an instant, personalised itinerary. Being able to draw out the overall trending mood expressed across the cacophony of social media channels (tweets, blogs etc) is very exciting but also quite challenging. The recognised powerhouse of social media analytics, Radian 6, blogged that “There will always be a need for human review and involvement to verify automated results, and ensure that sentiment levels are tagged within the context of individual and unique business goals and agreed upon criteria.” (emphasis added)

To be fair, the Radian 6 blog was posted eons ago in Dec 2009 but there has not been a follow-up since to instill more confidence in the automated sentiment technology offering. A subsequent post in 2011 walks the user through the tool but does not touch on how much weight should be given to these results.

Social Eyez is another heavy hitter in the social media analytic space with a focus on the Middle East North Africa (MENA) region. The promising claim that “..Socialeyez effectively identifies, monitors, and analyzes the most important and relevant social media content in virtually all languages, including Arabic – perhaps the most linguistically complex language in the world.” seems to be the holy grail but I missed the first part of that sentence “By using human experts..“. Social Eyez notes further that “Technology plays a key role in our everyday lives, and is key to the success of SocialEyez. However, we do not believe that today’s technology has reached the levels where a fully automated report can offer accurate and truly actionable insight.

Automated analysis of social media sentiment is certainly not easy!

Drawing on my personal experience in enterprise BI and reporting however, the goal is not 100% accuracy. Rather, it is the capability to discern an overall sentiment & trend and associate this with a level of confidence. According to comparative review by toptenreviews, Sysomos is one tool that stands out with a perfect 10 for Sentiment Accuracy (ahead of Radian 6 at 9.4).


Unfortunately, the demo seems to be for enterprises only so I could not personally review Sysomos. The samples provided seem promising however with the screenshot below from recent news:

Then of course, after NLP is mostly solved, the next challenge will be to be convincing.. http://bit.ly/10qMpOz


Leave a comment

Filed under Analytics

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s