Anomaly Detection

name: Anomaly Detection
description: >-
  A curated collection of APIs, tools, and platforms for detecting anomalies in
  data streams, time series, and multivariate metrics. Covers cloud ML services,
  observability platforms, and open-source frameworks used for fraud detection,
  predictive maintenance, IoT monitoring, and security analytics.
image: https://kinlane-productions2.s3.amazonaws.com/apis-json/apis-json-logo.jpg
created: 2024-01-15 00:00:00+00:00
modified: 2026-04-19 00:00:00+00:00
specificationVersion: '0.16'
url: https://raw.githubusercontent.com/api-evangelist/anomaly-detection/refs/heads/main/apis.yml
apis:
  - name: Azure AI Anomaly Detector
    description: >-
      Azure AI Anomaly Detector is a managed REST API service that enables
      monitoring and detection of anomalies in time series data without requiring
      machine learning expertise. Supports univariate batch and streaming
      detection, multivariate detection using Graph Attention Networks for up to
      300 correlated signals, and change-point detection. The service is being
      retired on 1 October 2026 in favor of Microsoft Fabric real-time
      intelligence.
    image: https://kinlane-productions2.s3.amazonaws.com/apis-json/apis-json-logo.jpg
    humanURL: https://learn.microsoft.com/en-us/azure/ai-services/anomaly-detector/overview
    baseURL: https://api.cognitive.microsoft.com
    tags:
      - Anomaly Detection
      - Azure
      - Machine Learning
      - Microsoft
      - Multivariate
      - Time Series
      - Univariate
    properties:
      - type: Documentation
        url: https://learn.microsoft.com/en-us/azure/ai-services/anomaly-detector/overview
      - type: APIReference
        url: https://learn.microsoft.com/en-us/rest/api/anomalydetector/
      - type: Quickstart
        url: https://learn.microsoft.com/en-us/azure/ai-services/anomaly-detector/quickstarts/client-libraries
      - type: Tutorials
        url: https://learn.microsoft.com/en-us/azure/ai-services/anomaly-detector/tutorials/batch-anomaly-detection-powerbi
      - type: GitHubRepository
        url: https://github.com/microsoft/anomaly-detector
    contact:
      - FN: Microsoft Azure Support
        url: https://azure.microsoft.com/en-us/support/

  - name: Elasticsearch Anomaly Detection API
    description: >-
      Elasticsearch Machine Learning APIs provide a comprehensive suite of
      anomaly detection capabilities for time series data stored in Elasticsearch
      indices. Supports creating and managing anomaly detection jobs and
      datafeeds, accessing bucket, record, category, and influencer results,
      model snapshots, calendars, scheduled events, and forecasting. Part of the
      Elastic Stack ML feature set available in subscriptions.
    image: https://kinlane-productions2.s3.amazonaws.com/apis-json/apis-json-logo.jpg
    humanURL: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-apis.html
    baseURL: https://your-elasticsearch-host:9200
    tags:
      - Anomaly Detection
      - Elasticsearch
      - Machine Learning
      - Monitoring
      - Time Series
    properties:
      - type: Documentation
        url: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-apis.html
      - type: APIReference
        url: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-ad-apis.html
      - type: GettingStarted
        url: https://www.elastic.co/guide/en/machine-learning/current/ml-ad-overview.html
      - type: GitHubOrganization
        url: https://github.com/elastic
    contact:
      - FN: Elastic Support
        url: https://www.elastic.co/support

  - name: Datadog Anomaly Monitor API
    description: >-
      Datadog's Monitors API supports anomaly detection monitors that identify
      unusual metric behavior using historical pattern analysis including trends,
      day-of-week, and time-of-day seasonality. Offers three detection
      algorithms — Basic, Agile (SARIMA), and Robust (seasonal-trend
      decomposition) — configurable via REST API. Available across regional
      endpoints for US, EU, AP1, AP2, GOV, US3, and US5 deployments.
    image: https://kinlane-productions2.s3.amazonaws.com/apis-json/apis-json-logo.jpg
    humanURL: https://docs.datadoghq.com/monitors/types/anomaly/
    baseURL: https://api.datadoghq.com
    tags:
      - Anomaly Detection
      - Datadog
      - Monitoring
      - Observability
      - Time Series
    properties:
      - type: Documentation
        url: https://docs.datadoghq.com/monitors/types/anomaly/
      - type: APIReference
        url: https://docs.datadoghq.com/api/latest/monitors/
      - type: Authentication
        url: https://docs.datadoghq.com/api/latest/authentication/
      - type: GitHubOrganization
        url: https://github.com/DataDog
    contact:
      - FN: Datadog Support
        url: https://www.datadoghq.com/support/

  - name: AWS Lookout for Metrics
    description: >-
      Amazon Lookout for Metrics is a fully managed ML service that
      automatically detects anomalies in business and operational data. It
      connects to data sources including Amazon S3, Amazon Redshift, Amazon
      CloudWatch, and SaaS applications, learns each metric's normal behavior,
      and sends alerts when anomalies are detected. Provides root cause analysis
      grouping related anomalies for faster diagnosis.
    image: https://kinlane-productions2.s3.amazonaws.com/apis-json/apis-json-logo.jpg
    humanURL: https://aws.amazon.com/lookout-for-metrics/
    baseURL: https://lookoutmetrics.us-east-1.amazonaws.com
    tags:
      - Amazon Web Services
      - Anomaly Detection
      - AWS
      - Business Metrics
      - Machine Learning
    properties:
      - type: Documentation
        url: https://docs.aws.amazon.com/lookoutmetrics/latest/dev/lookoutmetrics-welcome.html
      - type: APIReference
        url: https://docs.aws.amazon.com/lookoutmetrics/latest/api/Welcome.html
      - type: GettingStarted
        url: https://docs.aws.amazon.com/lookoutmetrics/latest/dev/lookoutmetrics-gettingstarted.html
      - type: Pricing
        url: https://aws.amazon.com/lookout-for-metrics/pricing/
    contact:
      - FN: AWS Support
        url: https://aws.amazon.com/contact-us/

  - name: PyOD (Python Outlier Detection)
    description: >-
      PyOD is a comprehensive and scalable Python library for detecting
      outliers/anomalies in multivariate data. It includes more than 40
      detection algorithms including deep learning approaches (AutoEncoder,
      VAE), proximity-based methods (LOF, CBLOF), linear models (PCA, OCSVM),
      and ensemble methods (IForest, LOCI). Widely used in research and
      production for fraud detection, intrusion detection, medical anomaly
      detection, and data quality monitoring.
    image: https://kinlane-productions2.s3.amazonaws.com/apis-json/apis-json-logo.jpg
    humanURL: https://pyod.readthedocs.io/
    baseURL: https://pypi.org/project/pyod/
    tags:
      - Anomaly Detection
      - Data Science
      - Machine Learning
      - Open Source
      - Outlier Detection
      - Python
    properties:
      - type: Documentation
        url: https://pyod.readthedocs.io/en/latest/
      - type: APIReference
        url: https://pyod.readthedocs.io/en/latest/pyod.html
      - type: GitHubRepository
        url: https://github.com/yzhao062/pyod
      - type: SDK
        url: https://pypi.org/project/pyod/
    contact:
      - FN: PyOD Maintainers
        url: https://github.com/yzhao062/pyod/issues

maintainers:
  - FN: Kin Lane
    email: [email protected]
    X: apievangelist
    url: https://apievangelist.com
tags:
  - Anomaly Detection
  - Artificial Intelligence
  - Data Science
  - Fraud Detection
  - Machine Learning
  - Monitoring
  - Observability
  - Outlier Detection
  - Pattern Recognition
  - Security
  - Time Series
include: []
common:
  - type: GitHubOrganization
    url: https://github.com/api-evangelist/anomaly-detection
  - type: BestPractices
    url: https://pyod.readthedocs.io/en/latest/faq.html
  - type: Blog
    url: https://techcommunity.microsoft.com/t5/AI-Customer-Engineering-Team/Introducing-Azure-Anomaly-Detector-API/ba-p/490162
  - type: JSONSchema
    url: https://raw.githubusercontent.com/api-evangelist/anomaly-detection/refs/heads/main/json-schema/anomaly-detection-anomaly-schema.json
    title: Anomaly Schema
  - type: JSONSchema
    url: https://raw.githubusercontent.com/api-evangelist/anomaly-detection/refs/heads/main/json-schema/anomaly-detection-time-series-schema.json
    title: Time Series Schema
  - type: JSONSchema
    url: https://raw.githubusercontent.com/api-evangelist/anomaly-detection/refs/heads/main/json-schema/anomaly-detection-detection-job-schema.json
    title: Detection Job Schema
  - type: Vocabulary
    url: https://raw.githubusercontent.com/api-evangelist/anomaly-detection/refs/heads/main/vocabulary/anomaly-detection-vocabulary.yaml
  - type: Features
    data:
      - name: Univariate Time Series Detection
        description: >-
          Detect anomalies in a single time series metric using statistical
          algorithms, SARIMA models, and SR-CNN approaches for both batch and
          real-time streaming use cases.
      - name: Multivariate Detection
        description: >-
          Identify anomalies across multiple correlated metrics simultaneously
          using graph attention networks and correlation analysis, capturing
          system-level failures invisible in individual metrics.
      - name: Streaming and Batch Modes
        description: >-
          Support for both real-time streaming anomaly detection on incoming
          data points and batch retrospective analysis across historical
          datasets.
      - name: Change Point Detection
        description: >-
          Identify structural breaks and trend changes in time series data
          beyond point anomalies, enabling detection of regime shifts and
          concept drift.
      - name: Root Cause Analysis
        description: >-
          Group related anomalies and surface likely contributing factors to
          accelerate diagnosis and response.
      - name: Algorithm Diversity
        description: >-
          Access to a wide range of detection algorithms from statistical
          methods to deep learning, including IForest, LOF, OCSVM, AutoEncoder,
          VAE, and SARIMA.
  - type: UseCases
    data:
      - name: Fraud Detection
        description: >-
          Identify fraudulent transactions, account takeovers, and suspicious
          behavioral patterns in financial and e-commerce systems.
      - name: Predictive Maintenance
        description: >-
          Detect early signs of equipment failure in industrial IoT systems by
          identifying anomalous sensor readings before breakdowns occur.
      - name: IT and Security Operations
        description: >-
          Detect unusual network traffic, unauthorized access patterns, and
          security incidents in real time using behavioral baselines.
      - name: Business Metrics Monitoring
        description: >-
          Alert on unexpected drops or spikes in KPIs such as revenue,
          conversion rates, user engagement, or API error rates.
      - name: Healthcare Monitoring
        description: >-
          Monitor patient vitals, lab values, and medical device readings for
          out-of-range or clinically significant anomalies.
  - type: Integrations
    data:
      - name: Amazon S3
        description: >-
          Connect anomaly detection pipelines to S3 data lakes for batch
          analysis of historical metric data.
      - name: Elasticsearch / OpenSearch
        description: >-
          Use Elasticsearch ML datafeeds to continuously analyze indices for
          anomalous patterns using built-in anomaly detection jobs.
      - name: Amazon CloudWatch
        description: >-
          Pipe CloudWatch metrics into AWS Lookout for Metrics for automated
          operational anomaly alerting.
      - name: Microsoft Fabric / Real-Time Intelligence
        description: >-
          Migration target for Azure Anomaly Detector users, providing
          integrated real-time anomaly detection within the Microsoft Fabric
          analytics platform.
      - name: Grafana
        description: >-
          Visualize anomaly scores and detected anomalies from Elasticsearch ML
          and Datadog within Grafana dashboards.