Coniferest: a complete active anomaly detection framework

TL;DR

Coniferest is an open-source Python framework for active anomaly detection, supporting static and active outlier detection algorithms, evaluated on synthetic and astronomical datasets.

Contribution

It introduces a comprehensive, open-source framework with multiple algorithms for active anomaly detection, including real-world application examples.

Findings

Effective detection on synthetic datasets

Successful application in astronomical data analysis

Open-source tool supporting multiple algorithms

Abstract

We present coniferest, an open source generic purpose active anomaly detection framework written in Python. The package design and implemented algorithms are described. Currently, static outlier detection analysis is supported via the Isolation forest algorithm. Moreover, Active Anomaly Discovery (AAD) and Pineforest algorithms are available to tackle active anomaly detection problems. The algorithms and package performance are evaluated on a series of synthetic datasets. We also describe a few success cases which resulted from applying the package to real astronomical data in active anomaly detection tasks within the SNAD project.