LinkedIn Details Efforts to Stamp Out Fake Accounts in New Report
While Facebook and Twitter have both detailed their expanding efforts to detect and remove fake profiles, and improve the integrity of their respective platforms, LinkedIn has also been working hard on the same, with many ill-intentioned groups also seeking to infiltrate the professional social network.
In fact, fake profiles on LinkedIn are likely a bigger problem than you'd imagine - according to LinkedIn, their processes recently blocked five million suspicious accounts from being created within a single day.
When you think about it, it actually makes a lot of sense - LinkedIn is where you post your job history and contact information, and given that many people use LinkedIn to source business leads, the contact info on their LinkedIn profiles is probably more likely to be up to date than the same on other platforms.
When you add a new connection on LinkedIn, you give them access to all of this information, which can make data grabbing on the platform particularly effective.
That's why LinkedIn's working to stop it, and protect its users - the aforementioned five million accounts were blocked as part of LinkedIn's advanced detection process, which works through both automation and human review.
As explained by LinkedIn:
"One of the ways we maintain a safe and trusted professional community on LinkedIn is by requiring that every LinkedIn profile must uniquely represent a real person. One of the ways we ensure that accounts are real is by building automated fake account detection systems at scale for detecting and taking action against fake accounts. These allow us to protect our members from bad activity by bad actors."
LinkedIn's detection process is detailed in this graphic.
Essentially, LinkedIn's systems have certain behaviors and traits they look for in order to highlight potentially fake accounts.
One of those measures is volume - as noted by LinkedIn:
"For many types of abuse, attackers require a large number of fake accounts for the attack to be financially feasible. Thus, in order to proactively stop fake accounts at scale, we have machine-learned models to detect groups of accounts that look or act similarly, which implies they were created or controlled by the same bad actor."
That's how LinkedIn's able to detect and block large numbers of accounts at once.
In terms of new users signing up, LinkedIn says each new registrant is assessed by its system:
"Every new user registration attempt is evaluated by a machine-learned model that gives an abuse risk score. Signup attempts with a low abuse risk score are allowed to register right away, while attempts with a high abuse risk score are prevented from creating an account. Attempts with medium risk scores are challenged by our security measures to verify that they are real people. This registration model is quite effective at preventing bulk fake account creation."
Outside of this, LinkedIn's systems can also detect suspect behavior by looking at what an account does. If a new account shares a heap of content from a single source, repeatedly shares the same link, sends out a heap of connection requests - all of these variables can be assessed, in some form, by LinkedIn's detection systems.
"We have many models that either look for specific types of bad behavior typical to abusive accounts or behavior that is anomalous. Additionally, our systems have redundancy, which ensures that fake accounts not caught by the early stages of our defenses (top of the funnel) are eventually caught by the latter ones (bottom of the funnel)."
On top of this, LinkedIn has human reviewers and assessors evaluating activity, and looking for patterns and concerns.
As noted, while Facebook and Twitter are rightfully at the forefront of the investigation into how the proliferation of fake profiles on social networks has helped increase the spread of fake news - and potentially influenced election outcomes - other platforms also have valuable data, and are also regularly the subject of similar attacks.
LinkedIn, in particular, has a wide range of valuable data insights, so its good to know that the platform's working to secure its systems and protect its users.
You can read the full report into LinkedIn's automated fake account detection measures here.
Follow Andrew Hutchinson on Twitter