Debate Heats Up on Sentiment Analytics


Remain calm!

If you do and another million Twitter users do the same, it can tell us a lot about what the stock market will do in the next couple of days. If you don't stay calm, it may tell us just as much.

Some insist the emotional state or feelings of users on vast social media networks like Twitter and Facebook (and now potentially Google+) are measurable and therefore useful for those trying to predict trends and events -- a process called sentiment analysis. Using semantic Web technology, this process determines the emotional state of communicators based on the valuation of words often ranked for basic emotional context.

But not everybody believes our ability to read these signals are advanced enough to make them useful in analytical models for marketers, investors, and others who might find the information helpful. Social media consultant Joe Stanganelli and Web analytics consultant Pierre DeBois recently squared off on the "sentimental value" of social media data in a Point/Counterpoint debate. Yesterday they ramped up the discussion in an instant e-chat with the AllAnalytics.com community.

"We can howl all day about how it shouldn't work, how stupid it is, and how there are all sorts of problems with it -- but time after time it's been proven by people who are way smarter than us," Stanganelli argued during Thursday's chat.

"I think it's early in the game to declare sentiment analysis as being an answer to marketers' prayers," DeBois maintained. The technique's effectiveness is too unproven for now, and "many businesses are still struggling with interpretation of data and how to incorporate social media into a strategy at the get-go."

What's more, DeBois said, measurement is still linked to basic Web architecture. As that architecture changes to address how we engage the Web today, the measurement process gets even trickier. Measurement "can be misinterpreted or overemphasized without context."

During the chat, AllAnalytics.com community member Cordell Wise sided with DeBois, citing his concern over selection bias. "Just because a bunch of people on [Facebook] launch a campaign to bring back butter pecan [at a local ice cream shop] doesn't mean it's a widely held view."

However, that doesn't mean the company should dismiss this data, Wise said, since it's still valuable. "Rather than looking for something that points to a broad sentiment, maybe it's more useful for spotting things like oversights or gauging reactions you didn't expect -- outliers, so to speak."

But Stanganelli has pointed to a growing body of research, including a study showing analysts' uncanny ability to predict stock market performance based on sentiments expressed on Twitter and another showing equally amazing accuracy in predicting motion picture box office receipts from Tweets. He insists that marketers and data miners should not disregard an application of such clear promise.

"That’s not to say we shouldn't still keep doing what we can to… hammer out the kinks, but there really is something here that can change the world," Stanganelli said.

You can read the entire chat exchange here, and you can share your thoughts on sentiment analysis. Can social media predict the mood of users across the planet, and can this data be mined and used for marketing or other purposes? How would your business or organization use such data if it became available? Please leave your comments and start a discussion below.

Shawn Hessinger, Community Editor

Shawn Hessinger is a community manager, blogger, social media and tech enthusiast, journalist, and entrepreneur based in Northeastern Pennsylvania. He serves as community manager and blogger for BizSugar.com, a business news and information Website, and contributes regularly to the online business news source, Small Business Trends. He is the founder of PostRanger.com, an online content and media community, and has provided blogging and social media services and consulting for companies all over the world. He researches and writes on a variety of business, Internet-related, and other tech topics including business intelligence and analytics. He is also keenly interested in computer-aided data management as it relates to his various online ventures. A newspaper journalist with more than 11 years experience as a reporter and then managing editor, Shawn began blogging in 2006 and now provides a variety of consulting and outsourcing services in Search Engine Optimization, Web development, and online marketing to companies large and small. He is a strong advocate for the use of BI and related computer data management in business decision making, whether using software as a service (SaaS), cloud, or other applications, and in the opportunity these technologies provide to transform small startups and larger established businesses alike.

BCBSNC, SAS Team on Advanced Analytics

The key to improving heathcare outcomes is to look at individual needs, the companies say.

Spoofing, Privacy Greatest Barriers for Biometrics

In Wednesday's e-chat, we discussed the analytics of identification and whether the technology might find a bigger role one day in marketing intelligence.


Re: Gaming the system
  • 8/31/2011 10:22:08 PM
NO RATINGS

What a great quote " behavior online may closely mirror  behavior in the real world". 

There is so much correlation that could be made against hvaing so much raw data that could be applied to so many things (disasters, holidays, etc) to see how modes changes and how they in turn effect other situations. 

Very cool stuff.

Re: Gaming the system
  • 8/31/2011 10:19:22 PM
NO RATINGS

Great article!! I'm sorry I missed the chat, I'm going to go through the chat and review the converstation.

There are really some untold metrics that could be harnessed from this data. This is really exciting work.

Re: sentiment analysis
  • 8/28/2011 9:23:37 PM
NO RATINGS

Yes, Klout has a long way to go, but once tools like it become more refined -- or enterprises can design proprietary tools that are similar to measure social media influence by individuals -- then they will be useful. Klout appears now to lean too heavily on what's going on in Twitter and has no clue how to measure LinkedIn.

Re: sentiment analysis
  • 8/25/2011 10:37:15 PM
NO RATINGS

Are you talking about Klout for measuring social media influence? I think it's still too early for them to claim what they do. According to their metric, I'm influential about photography and iPad (but I'm not really). I look forward to better tools to measure influence in the future.

Re: sentiment analysis
  • 8/23/2011 10:47:41 PM
NO RATINGS

Shawn, maybe I am being too simplistic here, but could you not measure how many referrals you're getting from that blogger as a percentage of your usual and customary web traffic. And they now have tools to measure individual's social media influence--these will only get more refined and "accurate." Combine the two and you have some sort of measurement.

Sentiment analysis
  • 8/23/2011 2:40:24 AM
NO RATINGS

Shawn it’s interesting that you ask the question about tweets to registers ringing because that is the fundamental issue with so many social media campaigns. I believe the social media activity needs to be tracked definitively with specific actions that are required. Companies should define their social media activities based on the specific activity  they are trying to initiate; feedback, promote a new product, offer a promotion, and create an action to the social media effort that is trackable.

Re: sentiment analysis
  • 8/22/2011 11:37:10 PM
NO RATINGS

Beth,

I'm wondering if your friends at the dairy have done any experimentation on the possible correlation between the number of brand mentions in social media like Twitter and Facebook and total earnings. The focus Starbucks and some other large brands have put on social media in recent years suggests there is an underlying knowledge of what social media may be doing for the brand. This, in turn, makes me wonder if some kind of analytics model has been developed by brands like Starbucks that really clearly reveals an ROI for the branding power social media can offer. 

Re: sentiment analysis
  • 8/22/2011 11:14:41 PM
NO RATINGS

Hi Broadway,
Yep, you're right about options for ROI. The trouble is there are uses for social media which marketers know anecdotally are effective but are nearly impossible to measure. Take the SEO effect of social media links on a Website. Facebook and Twitter links carry no link juice. Yes, you can see the number of visitors dropping in from Twitter or Facebook with simple Web analytics tools, but how can you measure the number of those visitors who have their own blogs and create powerful links carrying link authority to your Website. These may be bloggers who would never have found your Website without Facebook or Twitter but how can this value be quantified.

Re: sentiment analysis
  • 8/22/2011 10:39:12 PM
NO RATINGS

Of course!  Indicators can point to either the past, present or future. 

Re: sentiment analysis
  • 8/22/2011 10:34:56 PM
NO RATINGS

This is a great point, Seth, but the thing to remember is that sentiment analytics is sometimes being used to predict the future so, in this instance, when you are deciding on the performance of the stock market in the future based on what Twitter says today there is no room for multiple interpretations and the only way to double check your results is by loosing a bundle on Wall Street.

Page 1 / 2   >   >>
INFORMATION RESOURCES
ANALYTICS IN ACTION
CARTERTOONS
VIEW ALL +
QUICK POLL
VIEW ALL +