Platt Perspective on Business and Technology

Using social media as job search and career development business analysis resources 18: adding Facebook to the mix

This is my eighteenth installment to a series on the points of intersection between business intelligence and its gathering, and social media and related interactive online channels as sources of actionable information and insight, as they can be applied to job search and career development (see Guide to Effective Job Search and Career Development – 3, postings 397 and following for Parts 1-17.)

Up until here I have focused on a range of social media tools as they can be used and both in order to learn more about possible places of employment (Parts 1-8) and to better manage the online message that you share about yourself as a professional (Parts 9-17.) And I have focused, and certainly in a social networking context, on more explicitly professionally oriented sites such as LinkedIn and on resources where you can more directly manage and control the professional message that you would convey in all of this, and particularly for my discussion of Parts 9-17. And I have intentionally deferred discussion of Facebook in all of that, until here.

Facebook is by far the largest and most widely used social media site in current use on Earth. And a great many businesses of all types, and a great many individual professionals post there, and interactively respond to the postings of others there too. To put this claim into numerical perspective, Facebook as a social media business claims that it had approximately 1.49 billion user accounts in its system as of July, 2015 (and see some of the other statistics for that site and organization, reported there too.)

Facebook represents the largest and most diverse interactive social media forum in existence. And that last point specifically raises the challenge that Facebook presents to professionals who seek to present themselves effectively as such, online while still remaining socially connected to and involved online with their social acquaintances and friends too. Facebook is used by professionals but it is only peripherally business or career oriented or supportive. And it is the ongoing flood of all the rest that shows through that site that creates opportunity for mixed messages and for wrong messages, and certainly where they might be viewed out of context by a hiring manager, or by a current employer who wants to know more about their employees. And this flood of social media content includes photos as well as text and links to other resource sites. And it is their seemingly endlessly large trove of photos as social media content that I turn to next.

Many and perhaps even most active Facebook users post at least some photographs of themselves and their friends to their Facebook walls, and to those of their Facebook friends. And that often includes photos taken in social situations such as parties, where everyone in them is acting in ways that might make sense in the contexts that those photographs were taken but that could mislead if only viewed outside of those contexts. Social, social media content does not always positively reflect or support the professional image or online professional message of the people shown in it.

• If a hiring manager sees several photos of a possible new hire drinking and acting out in what looks to be a loud and even boisterously disruptive public setting, will that give them pause for thought about that potential hire’s judgment when making their hiring decision?
• What if this was an essentially one time only event at a best friend’s bachelor party, right before they got married? And what if no one involved in this was going to drive after drinking, or otherwise act irresponsibly and by pre-considered planning (none of which would be visible in those party photos)?

When you or someone else posts a photo on Facebook, it captures the images of people who were intentionally being photographed, but also of everyone else in the line of sight of your camera, who might not have even been involved in your party or event. This means that people can be photographically visible on Facebook without even knowing that – and without even knowing the people who have posted those images live to the site.

What does this mean? If no one has text tagged photos on Facebook that show you, to name you as being in them, and certainly for what might seem to be potentially compromising photographs, who would know of them or of you being in them? In principle, and lacking any compromising factors that would out you for specific photos (e.g. such as identifying text messages appended to the Facebook walls where those photos show), the sheer volume of Facebook profiles, walls and content should impose a measure of confidentiality by what would sound like highly secure steganographic means – burying those images under the flood of everything else on the site. But is this sense of security that these photos would not be found by a hiring manager or current employer sound?

To address that question, and that presumption, let’s start by considering the numbers of photo files involved here, actively maintained on the Facebook servers. By September, 2013, Facebook users had already uploaded more than 250 billion photos to the site and were actively adding some 350 million more of them every single day. At that rate of increase, Facebook would have close to twice that total number of member-uploaded photos in its system by the end of July, 2015. But that is misleading as the per-day new photo accumulation count has in fact continued to trend upwards since 2013.

That in and of itself might suggest that any single untagged photo would represent a very tiny needle lost in a particularly vast haystack, effectively protecting most everyone involved by the sheer volume of surrounding unrelated image file content. But that is where automated data processing and facial, and more generally visual pattern recognition software enter this narrative. And in anticipation of my next series installment to come, this is where I write about where social media is heading as much as I do about where it is now, and both in how it is used socially and in how it is and will be used in more professional contexts.

Facebook holds in its servers, one of the largest and most comprehensive collections of photographic data on the planet, and certainly one of the largest that is privately held rather than governmental in nature. Facebook also actively pursues one of the largest and I add most effective automated facial recognition software programs on Earth, with its DeepFace program. See DeepFace: Closing the Gap to Human-Level Performance in Face Verification for a summary document on this program and link to the full text of a 2014 publication on it.

In early 2014 DeepFace was already able to correctly identify individuals photographed with an accuracy of just under 94%, and this was when these people were being photographed “in the wild” (e.g. without any staging such as positioning their faces so they were all image-captured from some specific angle, or in some specific preferred lighting.) That essentially matches human brain performance for facial recognition. And perhaps even more impressively, this system proved at that time to be able to correctly identify individuals with an 83% reliability when their face was not clearly enough visible to allow their identification from that. DeepFace can also identify individual people from image analysis of features such as their body proportions, and their clothing, their hair color, et cetera as analyzed in aggregate. And the accuracy of this system has continued to improve since then, along with the collateral supportive software systems needed to apply this image recognition technology in a big data image file context and to even vastly large database holdings.

We are rapidly approaching a situation where there can be no anonymity in numbers or in the volume of image data that a photo might be buried in. And in this context, I add that Facebook is not alone, and even when just considering private sector participants. Google has a tremendously active and far-reaching program for automated facial recognition from online and database stored image files too, even if it is one that is nowhere near as advanced as that of Facebook, at least as of this writing. Note: I have written so far in this posting, primarily in terms of still-photographic image analysis, but it is much easier to assemble the types of 3-D model representations that programs such as DeepFace use, when these tools are applied to video. Facebook, and Google and others are developing, and they seek to widely apply this type of technology to all image format types and with a goal of readily, quickly and accurately identifying anyone and everyone showing in their systems. And beyond that, their goal is to merge facial recognition and other image-based identification with a more comprehensive individual identification and analysis program that cuts across image, text and all other data types in assembling descriptive and predictive models.

• Who would get to see all of this, or at the very least see aggregate-model information that could be assembled from it, with those models becoming increasingly able to identify individuals and track them, no matter where or how they appear on a site such as Facebook?

Progressively finer-grained if still individually anonymous demographic modeling data, and demographic-of-one individualized data are becoming both increasingly possible and increasingly valuable in a marketing and sales context (see my series Big Data and the Assembly of Global Insight Out of Small Scale, Local and Micro-Local Data, at Reexamining the Fundamentals, as Topic Section IV of that directory.) And Facebook does provide large and I add ever increasing amounts and ranges of such data for a fee, to its business partners and to its business clients. And its list of clients has come to include hiring and employing businesses too, who seek as part of their critically important due diligence, to better and more fully know about the people who they might hire or who they already have. And it has come to include third party businesses too, that do this type of new hire candidate and employee background research for a fee, for other businesses.

I am going to turn in my next series installment, to consider the implications of this developing trend. And in the course of that I will examine, and in effect reexamine some of the fundamental assumptions that I have been making throughout this series, for when we all have to assume that our lives will become effectively transparent and certainly for our online lives, and when the privacy of anonymity can no longer be assumed.

Meanwhile, you can find this and related postings at my Guide to Effective Job Search and Career Development – 3 and at the first directory page and second, continuation page to this Guide. And you can find this and related material at Social Networking and Business and its continuation page too.

Addendum note: I finished writing this posting a few days ago, then heard a news update from Facebook that I have decided to add here. Today as of this writing, August 27, 2015, Mark Zuckerberg announced that for the first time ever, one billion people have used Facebook in a single day. I add that to my statistics of note as offered above in the main body of this posting. And while I am adding an update to this, I also make note here of a new and still emerging functional development at Facebook that I offer in the same way that I made note of their DeepFace program, as an indication of what is to come: their new artificial intelligence-based user assistant: M (which is still in its earliest public-test stage of development. See Facebook Launches M, Its Bold Answer to Siri and Cortana. Facebook is very large and very powerful and it is still growing and it is developing what will quite likely become the benchmarks for other social media sites to strive towards, and in a variety of directions. It is not by any means a “business or professional” social media resource, and certainly not as a primary priority – but it is a site and a medium that every professional has to be prepared to deal with and both for its positives and for its negatives.

Leave a Reply

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

You are commenting using your 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

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: