A new tool released on Tuesday by Twitter as open source can be used by developers to detect various types of anomalies.
AnomalyDetection is a package for R, the free software environment for statistical computing and graphics. Twitter has been using the tool to detect anomalies such as spikes caused by user engagement on the social media platform during breaking news, major sporting events and holidays.
From a security standpoint, AnomalyDetection can be utilized to detect activities associated with bots and spam, which may cause anomalies in the number of followers and favorites. Anomalies can also be detected in system metrics after the release of new software, Twitter said.
“An anomaly can be positive or negative. An example of a positive anomaly is a point-in-time increase in number of Tweets during the Super Bowl. An example of a negative anomaly is a point-in-time decrease in QPS (queries per second). Robust detection of positive anomalies serves a key role in efficient capacity planning. Detection of negative anomalies helps discover potential hardware and data collection issues,” Twitter software engineer Arun Kejariwal explained in a blog post.
The social media giant has released AnomalyDetection as open source to give the community the chance to contribute to improving the tool. The R package is available on GitHub.
AnomalyDetection is not the only tool released by Twitter as open source. In October, the company announced the availability of BreakoutDetection, a tool that’s designed to detect breakouts. Unlike anomalies, which are characterized by point-in-time anomalous data points, breakouts are represented by a ramp up from one steady state to another.
Over the past months, Twitter has put a lot of effort into making the platform as safe and secure as possible. In December, the company announced the release of improved tools for reporting harassment and abuse. In September, Twitter launched a bug bounty program via HackerOne, promising security experts a minimum of $140 for every vulnerability they report.
A study conducted last year by Trend Micro revealed that of 570 million analyzed tweets, 33 million (5.8%) contained links to malware, spam, phishing pages and other threats.
