Anomaly Detection and Forecasting That Learns From Every Write in InfluxDB

The blog post introduces three River-based machine learning plugins for InfluxDB 3, which enhance operational time series data processing by enabling real-time anomaly detection, adaptive profiling, and short-horizon forecasting directly within the database. These plugins utilize River, a Python library for online machine learning, to incrementally learn from streaming data without requiring additional infrastructure or data pipelines. The River Anomaly Detector plugin allows users to detect anomalies using rolling Z-scores, seasonal patterns, and drift detection, while the River Auto-Profiler optimizes anomaly detection settings by profiling numeric series and generating tuning recommendations. The River Forecaster applies online ARIMA-style modeling to predict future data points, facilitating proactive operational decision-making. By embedding these plugins into InfluxDB, users can seamlessly integrate machine learning tasks with time series data, enabling efficient querying and visualization without relying on external platforms.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.