Anomaly Detection and Typical Challenges with Time Series Data
Anomaly detection, popularly known as outlier detection is a data mining process that aims to discover unexpected events or rare items in data and to determine details about their occurrences. Anomaly detection in time series data brings its own challenges due to seasonality, trends and the need to more complex multivariate analysis to yield better results.
In this multi-part blog series, we will discuss key aspects of anomaly detection, typical challenges that we encounter in doing anomaly detection for time series data, and finally discuss approaches for doing multivariate anomaly detection.
What is Anomaly Detection
Automatic anomaly detection is a critical aspect in today’s IT world where the sheer volume of data makes it impossible to tag outliers manually. Anomaly Detection can be used in IT Operations to trigger prompt troubleshooting, help avoid loss in revenue, and maintain the reputation and branding of the company. To serve this purpose, large companies have built their own anomaly detection services to monitor their business, product and service health. Anomalies once detected send out alerts to the IT Operators to make timely decisions related to incidents. It is extremely important to detect an Anomaly to run the ITOps smoothly, predict future interruptions and therefore prevent any incident that might lead to data damage and unnecessary resource spent. Also, the anomalies need to be detected at the right time to yield best results and avoid any false alerts that mislead the operations.
With the advent of Artificial Intelligence and AIOps platforms, the need for manual monitoring of anomalies and alerts has drastically reduced. Earlier human intervention was required at every level, to monitor and identify the anomalies within the ITOps. With AI changing the face of technology, a machine is trained to understand the pattern over a specific time period to detect and predict any existing or future anomalies on its own, without any human intervention.
read more https://bit.ly/34uInxp
Originally Published by CloudFabrix Software Inc
