Business Intelligence & the Art of Investment


My recent blog, Finding the Right Words With Text Analytics, spurred an interesting conversation on the topic of measuring sentiment in social media for data-driven business decisions -- a growing trend among industries of all sorts.

As one example, Joe Stanganelli, a consultant in brand management and marketing and AllAnalytics.com community member, pointed us to this post on the spring launch of a Derwent Capital Markets hedge fund that predicts fluctuations in the stock market by monitoring Twitter.

"Gathering and analyzing sentiment on social media is important because it's getting the information you need directly from the horse's mouth -- that is, the general public," Stanganelli told me in an email exchange.

On the All Analytics message board, Stanganelli indicated he thought it interesting that for the hedge fund Derwent is evaluating everyday communications from 10 percent of Twitter's estimated 10 million daily users and not Tweets specific to the stock market or made by stock market experts.

Derwent reportedly based its approach on a recent scientific article titled, "Twitter mood predicts the stock market," written by Johan Bollen and Huina Mao, with the School of Informatics and Computing at Indiana University, Bloomington, and Xiao-Jun Zeng, with the School of Computer Science at the University of Manchester in the UK.

In the article, the scientists suggest that fluctuations in the stock market may be predicted by levels of calm in the average Tweet. They're not alone in this belief.

For example, researchers at the University of Illinois at Urbana-Champaign predicted S&P 500 performance based on anxious or worried feelings expressed over LiveJournal blogs, and researchers at a German institution discovered Tweet volume and bullishness was predictive of market returns, Stanganelli said. In this latter case, however, researchers only used stock-related Tweets.

Of course, using social media to gauge the sentiments of your audience is nothing new. Media marketers have done this for years using various channels to get a better feel for their followers and customers. They've even done so to develop relationships and connections with their customer bases to get an additional sense of the group dynamics.

What is new is the degree to which companies can reduce these sentiments to pure data, using it to predict specific outcomes. Also unusual is the use of data gathered from the public at large and not deliberately from customers or interested people with any real understanding of the business being measured.

To what degree can this type of analysis be adapted to other industries and with what level of success? Could the sentiments of the general user on the social Web hold the key to predicting the success or failure of other businesses or ventures some day? Tell us what you think on the board 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: Deep unease
  • 8/15/2011 12:44:23 PM
NO RATINGS

aaphil: Yes.

Re: Deep unease
  • 8/15/2011 12:41:25 PM
NO RATINGS

Joe,

I know this line does not really apply to finance but:

 "Twitter-based" hedge fund

Is that real?

Re: Deep unease
  • 8/15/2011 9:29:21 AM
NO RATINGS

Well, you got me there, Joe. Indeed, everybody IS self-centered on Twitter. I guess that is the point. ;-)

 

Re: Deep unease
  • 8/15/2011 9:23:37 AM
NO RATINGS

It is exactly those kind of "inconsequential" Tweets that contribute to the algorithm's 87.6 percent accuracy in predicting the market in study.  It's not the objective, universal importance of the singular Tweet that matters; it's the subjective well being (SWB) that matters.  That's the point of the theory.

Each Tweet has an SWB score that is taken into account in the analysis of 100 million Tweets per week that Derwent Capital Markets analyzes in operating its "Twitter-based" hedge fund.

(Incidentally, one doesn't have to have a requisite level of purchasing power to be indicative or representative of global mood and sentiment.)

And have you been on Twitter lately?  EVERYONE there is self-centered.  ;)

(Follow me on Twitter at @JoeStanganelli!)

Re: Deep unease
  • 8/15/2011 8:08:05 AM
NO RATINGS

Joe -- well, I think, too that part of my unease comes from having teenagers in the house who have recently glommed onto Twitter. If we could somehow eliminate the feelings they express in social media and leave the measurements to those with, say, real purchasing power (ie, the ability en masse to move a market) then I'd feel more comfortable about the concept.  Teens can Tweet all day long about how excited and happy, or conversely, how disappointed they are about life in general but does that really matter when we're looking at the stock market? They are a notoriously self-centered bunch, after all.

 

Re: Deep unease
  • 8/13/2011 4:40:23 AM
NO RATINGS

Thanks, Shawn.  And thanks to you guys for inviting me to do it.  I enjoyed delving into the issue.

Re: Deep unease
  • 8/13/2011 3:47:14 AM
NO RATINGS

Joe,

I think the notion of measuring anxiety, calmness or other simple emotional states translated through social media as a whole is probably the most sensible method of looking at sentiment in social media currently and, though I can see some problems in figuring out how to go about reliably measuring it, I believe that if it could be measured it would certainly be predictive. Nice job by both you and Pierre on the point/counterpoint, by the way. I have no doubt this will continue to be an important issue. 

Re: Possible correlations
  • 8/13/2011 3:31:59 AM
NO RATINGS

magneticnorth,

And this is presumably what the study above indicates...

Re: Possible correlations
  • 8/13/2011 3:28:17 AM
NO RATINGS

aaphil,

In the example given above with the stock market and Twitter, the study indicated sentiments measured on social media predicted stock market performance several days later. 

Re: Deep unease
  • 8/12/2011 11:31:39 PM
NO RATINGS

Hi, Beth.  I think those misgivings are much easier to get over when you forget the measurement tactics and look at the overall hypothesis being tested -- that the market is significantly affected by how calm, worried, or anxious people feel in their lives (whatever the reason or source for those feelings).

If you can accept that that's a worthwhile hypothesis to test and explore as quite possibly true, then the next step is about looking for ways to test it.

Right now, I can't think of a purer, more efficient, and more cost-efficient way of testing that than through the content people post on social networks.

In a vacuum, yes, the notion that someone's reactions to what happened on, say, Dance Moms the other night can affect the stock market in a meaningful way may sound farfetched -- but looking at the entirety of the situation in context, and considering humanity as a whole, it seems not only plausible, but likely.

(Incidentally, for what it's worth, I still remember how deeply I was affected and how morally outraged I was by the fourth season finale of The Apprentice.  Days later I was still thinking about it.  Pretty amazing that I was so deeply affected, I suppose, but feelings are feelings, and they no doubt had some degree -- however modest -- of influence on my behavior, as feelings tend to do.)

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