Method
Change-point detection
Detecting shifts in level or variance (CUSUM, PELT).
10 cases from 8 companies apply Change-point detection. 2 report results in production.
Typical tasks: Detect anomalies, Forecast demand and load, Catch fraud and abuse, Moderate content at scale and 2 more. Companies: Avito, Databricks, Walmart, VTB, Zillow and 3 more. Often combined with: Lag and window features, Gradient boosting, Backtesting, Time-series forecasting, Threshold selection and 35 more.
| Forecasting at Databricks: One Framework Behind Consumption and Infra Cost | saas-b2b | 2026-06 | |
| Moderation Platformization | classifieds | 2025-05 | |
| Autotuning: A comprehensive approach to Time Series Forecasting | retail | 2025-03 | |
| How We Cut Forecasting Time from Two Weeks to Five Minutes | classifieds | 2024-05 | |
| Time Series Segmentation: From Classical Approaches to Ensembles | general | 2024-05 | |
| ML in Predicting ESP Failures | — | manufacturing | 2024-05 |
| Anomaly Detection: Anomaly Scoring for Preventive Detection of Information System Failures | banking | 2024-05 | |
Product Features: The Key to Increasing Model Efficiency for Account Propensity![]() | ![]() | saas-b2b | 2023-10 |
| Preprocessing and Anomaly Detection in Time SeriesWT | WT | manufacturing | 2023-05 |
| SpectroBrain: Detecting Phone Spam with Semi-Supervised Learning | marketplace | 2023-05 |
