To keep you up-to-date with what’s going on in anomaly detection, we keep an ongoing list of the biggest glitches happening in the business world. Here is what made waves in June. June 25, 2019 When Dutch telco KPN suffered a major outage on the evening of Tuesday, June 25, the 112 emergency number was also knocked out across the country. “We have no reason to think it was (a hack) and we monitor our systems 24/7,” the company spokesperson told Reuters.
Companies invest in anomaly detection in order to proactively identify risks, such as revenue loss, customer churn and operational performance issues. Anomaly detection essentially enhances traditional BI and visualization tools, venturing beyond a summary view of your data. It constantly scans every metric, at a granular level, to find abnormalities. But in order for this technology to have an impact, you must be able to trust it.
Machine Learning (ML) algorithms are designed to automatically build mathematical models using sample data to make decisions. Rather than use specific instructions, they rely on patterns and inference instead. And the business applications abound. In recent years, companies such Google and Facebook have found ways to use ML to utilize the massive amounts of data they have for more profit.
Many data-driven companies struggle with tracking anomalous behavior in their KPIs, let alone their more granular metrics, because traditional BI tools just can’t keep pace with big data. Manual processes, whether monitoring dashboards or setting static thresholds, often leads to missed incidents or prolongs time to resolution.
Getting the right metrics at the right time could be the difference between running a smooth operation and making costly mistakes. And yet, data delays are still very much a part of data analysis today.
The global data supply reached 4.4 zettabytes (ZB) in 2013 – or 4.4 trillion GB – but less than five percent of that data was actually analyzed, according to the seventh Digital Universe report by the IDC. Approximately 22 percent of all existing data would actually be useful if tagged and analyzed.
5G is here and is widely expected to be a transformative communications technology for the next decade. This new data network will enable never-before-seen data transfer speeds and high-performance remote computing capabilities. Such vast, fast networks will need dedicated tools and practices to be managed, including AI and machine learning processes that will ensure efficient management of network resources and flexibility to meet user demands.
