Social networking on the World Wide Web has briskly matured to the point where it is a vital part of daily life. If you have followed my musings regularly here at Dashboard Insight, you will know that I have been a big proponent of using social networking applications for achieving better personal and business visibility and attaining day-in day-out productivity across a wide range of home and work responsibilities.
As we become more reliant on online social networking sites to connect with our friends, peers and business associates, we sometimes lose site of the fact we are generating an enormous amount of data about ourselves, potentially providing a treasure trove of demographics, opinions and behavioral information for third-party entities—from law enforcement agencies striving to track criminal movements to advertising companies looking for insight into consumer behavior.
Today, massive business intelligence (BI) efforts are being directed at social networking sites, as data mining organizations that sell demographic information and conduct opinion research are concentrating a huge amount of their energies on Facebook, Twitter, MySpace and others. These websites have become ground zero for their most important data mining efforts. (Ironically, while some sites such as Twitter are unable to show a profit or revenue stream, BI applications and companies that mine the data that is resident on the site are making some very tidy profits.)
There are innumerable data mining and data analysis techniques that can be successfully applied to social networking BI. Advanced techniques of pattern identification and classification (of interlocutors, topics and modes of conversation) provide statisticians and data miners with fundamental tools for unlocking the value of data strewn across a social networking site, or many sites. To make mention of a few of the more noteworthy techniques:
- Entity extractionlooks for trends in text, image and audio files. With a large enough sample size and a robust trending algorithm, distinct patterns of behavior will emerge; the means to predict future behavior will be enhanced. Entity extraction can also be used to better identify people, places, products, addresses, by assessing commonalities of highly prevalent characteristics. The concept of entity extraction is often co-mingled with that of pattern visualization, which attempts to model social collaboration by examining the linkages and social exchange between in-groups over time. Eventually coherent trend lines will emerge, along with clear leading and lagging indicators that describe the evolution of “group think” and person-to-person, person-to-group, and group-to-group communication.
- Clustering takes data items and groups them into related (and dissimilar) classes or taxonomies. Clustering can help identify customers and demographic groups that consume or buy products in similar ways.
- Association Rule Miningprovides predictive intelligence for data miners of social networking sites. It derives rules (that help predict future results) by utilizing pattern visualization (see above) to anticipate interaction between a site’s members - across the entire site and also within their smaller networking groups. For consumer products manufacturers, effective rules will be grounded in an understanding of how both good and bad opinions of a product are vocalized and passed on between a group’s members, or just as importantly, to members outside of a particular group of interest.
Despite a number of proven data mining techniques, extracting actionable business intelligence from social networks will remain a challenging endeavor; it will continue to be far from a cut-and-dried science. One of the main reasons for this is that people often represent themselves with fake names and false birthdates, or list phantom employers and addresses on their public profiles. Just about all frequent internet users have multiple online identities. (How many of us would not admit to having multiple profiles and personalities?) Such antics can make it quite difficult to get reliable demographic information from the internet.
Mass data mining of social networks is just the tip of the iceberg when it comes to how corporations are leveraging the information found on the internet. For instance, a large number of business organizations are now utilizing this data to assist in their due diligence efforts when hiring employees or choosing vendors. As a result, regular users of social networking sites such as MySpace or Facebook need to understand that the content they create on these sites may be used for much more than innocuous market research. Firms of all types have begun to partake in more creative background checks, e.g., combing the internet and social networks for more off-the-record information about people of interest. An off-color political comment or viewpoint or a risqué image posted at Flickr (a popular site for creating personal photo albums and uploading pictures) may cost some people a job in the future. This means that the days of casual posting to social networks are nearing their end for some of us.
It comes as no surprise that more than a few business executives I know have been abandoning Facebook and purging their profiles there permanently. For them, the risk of somebody writing things of questionable moral character on their profile wall far exceeds any perceived value. They are surrendering their online presence on certain sites in order to counteract a lack of control over activity streams and unwanted linkages.
For better or worse, if you are a participant on one the World Wide Web’s social networks, odds are that the content of your posts are being harvested to support massive BI and data mining schemes. And not just by big business having a look: local police departments are closely following what transpires on YouTube and Twitter; insurance companies are looking for travel pictures and posts of folks that are putatively (but not really) injured and collecting disability payments; federal governments and law enforcement agencies are scouring the data residing on social networks and job boards in hostile countries in order to gauge which people may pose a clear and present danger for terrorism. From this point on in time, social networking over the internet and cutting-edge business intelligence practices are close partners. Post your vacation travelogue or bachelor party pictures at your own peril.
About the Author
William Laurent is one of the world's leading experts in information strategy and governance. For 20 years, he has advised numerous businesses and governments on technology strategy, performance management, and best practices�across all market sectors. William currently runs an independent consulting company that bears his name. In addition, he frequently teaches classes, publishes books and magazine articles, and lectures on various technology and business topics worldwide. As Senior Contributing Author for Dashboard Insight, he would enjoy your comments at firstname.lastname@example.org
Copyright 2010 - Dashboard Insight - All Rights Reserved.