How Distil Networks uses machine learning to hunts down 'bad bots'

Source: How Distil Networks uses machine learning to hunts down 'bad bots'

robots.jpg

Image: iStockphoto/pxel66

Over the past few years, bots have started taking over parts of the tech world, with good bots like web crawlers indexing your site to boost your traffic and chat bots helping with more efficient communication in the office. Unfortunately, malicious bots are also on the rise, exposing vulnerabilities and stealing information.

In fact, 2014 was the first year that bots were purported to have outnumbered actual people online. With the exponential rise of bad bots, that’s a real problem. Distil Networks, a company that provides bot detection and mitigation services, recently raised $21 million in a Series C financing round to boost its efforts against bad bots.

For those unfamiliar, a bot is simply a piece of software that runs automated scripts online. Distil Networks uses machine learning algorithms to defend against the malicious behavior that can come from bots, such as web scraping, fraud, security breaches, spam, and downtime.

Distil offers an appliance product and a cloud CDN. Essentially, the product works by proactively detecting behavioral anomalies relative to the typical traffic patterns on your company’s website. The company also maintains a large database of known bad bots, and builds a unique “fingerprint” for the connecting browser so that, even if it tries to reconnect from another IP address or hide behind a proxy, the company will still be able to detect it.

SEE: Information security policy template (Tech Pro Research)

Distil Networks also offers products for threat intelligence and API security. After this latest round, the company’s total funding is $65 million, and it plans to put the money to use on global marketing and sales, as well as doubling its current headcount.

While the anti-bot market isn’t as developed as the antivirus space, Distil Networks isn’t alone. There are similar products offered by companies like Check Point Software, and Norton. And, bad bots aren’t just growing in number, they are growing in sophistication, too.

Many bots now meet the criteria for what is known as an Advanced Persistent Bot (APB), which are better at acting like humans online. According to a report by Distil, some 88% of bad bots could be classified as an APB.

According to a press release announcing the funding: “Their persistency aspect comes from their process for evading detection. For example, an APB might use 1000 IP addresses to make one request each, instead of one IP address to make 1000 requests, rendering impotent IP-centric defenses.”

Interested businesses can sign up for a free trial at the Distil Networks website.

The 3 big takeaways for TechRepublic readers

  1. Distil Networks recently raised $21 million to expand its bad bot hunting efforts globally, and to increase headcount.
  2. Distil Networks uses machine learning to identify and mitigate potential bad bots, fingerprinting them so that they can still be tracked if they reconnect from a different IP address.
  3. More and more bots are acting as Advanced Persistent Bots, performing more like humans online.

Also see