Anodot\'s New Patent Enables Accurate and Instant Insights, Furthering the Company\'s Vision of Completely Autonomous Business Operations

26 June 2019

REDWOOD, California, June 26, 2019 /PRNewswire/ -- Anodot, the Autonomous Analytics company, today announced that the United States Patent and Trademark Office is in the process of granting the company a new patent, which enables Anodot customers to receive alerts on business incidents that are both accurate and in real time.

Anodot's new filed patent titled System and Method for Efficient Estimation of High Cardinality Time-Series benefits Anodot customers through reducing estimation time of millions of time series metrics - a scale that is required for discovering business-related incidents. This ensures Anodot customers receive extremely accurate anomaly detection based on patterns for billions of time-series data measured continuously.

"This new patent is a natural progression in our vision to create fully autonomous business intelligence. With Anodot receiving millions of metrics each day, we needed to ensure both fast and precise detection of business incidents. This way, all of our clients' analyst teams can be alerted to and resolve issues immediately, without being occupied by false alerts," said Dr. Ira Cohen, Anodot's co-founder and Chief Data Scientist. "This patent guarantees the best service for our customers, providing accurate and instant insights each time."

Anodot's new patent follows two other patents that were granted recently, including System and Method for Transforming Observed Metrics into Detected and Scored Anomalies, (U.S. Patent 20160147583A1) and Fast Automated Detection of Seasonal Patterns in Time-Series Data Without Prior Knowledge of Seasonal Periodicity (US Patent 10061677B2), both recognizing the core underlying innovations that make Anodot the leader of Autonomous Analytics. Anodot currently has a fourth patent pending, Heuristic Inference of Topological Representation of Metric Relationships (US20160210556A1) which allows for the real-time detection of business incidents, finding issues in data before a human can. 

"Our new patent is a key achievement for Anodot. Being an Autonomous Analytics solution, we are furthering the ability for companies to become proactive, acting on issues in data before they even become apparent to consumers. As we look toward 2020, this patent is an essential component to our forecasting system, currently in beta, and is furthering our goal for completely autonomous business operations," said David Drai, CEO and Co-Founder of Anodot.

Anodot developed System and Method for Efficient Estimation of High Cardinality Time-Series as a system that estimates efficiently the normal patterns of time series data with long seasonal behaviors. State of the art prior methods for estimating the normal patterns in such cases is inefficient computationally. In an instance where a time series is measured every minute with a weekly seasonal pattern, state of the art methods can take up to 15 minutes in runtime, while Anodot's system cuts it down to 80 milliseconds - 3 orders of magnitude lower.

About Anodot

Anodot Autonomous Analytics is a machine learning platform that monitors and forecasts business performance. This turn-key solution independently understands behavioral patterns within times series data, to identify anomalies and to continuously anticipate future values. Anodot alerts operate in real time and offer context, correlating high-impact incidents to related factors, to dramatically reduce time to detection and resolution. Their pioneering technology is trusted by industry leaders across telco, eCommerce, fintech, adtech, gaming and more.

Anodot is headquartered in Silicon Valley and Israel, with sales offices worldwide. To learn more, visit us at and follow us on LinkedIn, Facebook and Twitter.

For more information, please contact:

Molly Meller 



Cision View original content: