impact.rst
1 PyOD Impact 2 =========== 3 4 Since its release in 2017, PyOD has become one of the most widely adopted anomaly detection libraries in the Python ecosystem. This page tracks external recognition: government and standards bodies, enterprise deployments, academic citations, books, courses, podcasts, and international tutorials. 5 6 For the full audit, see the `News & Media Coverage Audit <https://github.com/yzhao062/yzhao062.github.io/blob/main/news-coverage-audit.md>`_. 7 8 9 Government & Standards 10 ---------------------- 11 12 * **European Space Agency (ESA/ESOC)** implemented all 30 anomaly detection algorithms in the `OPS-SAT spacecraft telemetry benchmark <https://www.nature.com/articles/s41597-025-05035-3>`_ using PyOD 1.1.2. Published in *Nature Scientific Data* (2025). 13 14 15 Enterprise Deployments 16 ---------------------- 17 18 .. list-table:: 19 :widths: 20 80 20 :header-rows: 1 21 22 * - Company 23 - Usage 24 * - **Walmart** 25 - Real-time pricing anomaly detection, 1M+ daily updates (KDD 2019 industry paper) 26 * - **Databricks** 27 - Kakapo framework integrating PyOD with MLflow and Hyperopt 28 * - **Databricks** 29 - Insider threat risk detection solution using PyOD 30 * - **IQVIA** 31 - Healthcare fraud detection on 123K+ pharmacy claims using PyOD and SUOD models 32 * - **Ericsson** 33 - Patent `WO2023166515A1 <https://patents.google.com/patent/WO2023166515A1>`_ cites PyOD (Zhao et al., JMLR 2019) 34 * - **Altair AI Studio** 35 - Industry whitepaper using PyOD's Isolation Forest for anomaly detection 36 * - **Additional patents** 37 - `EP4662606A1 <https://patents.google.com/patent/EP4662606A1>`_ (EU), `CN111666198A <https://patents.google.com/patent/CN111666198A>`_ (China) both cite PyOD 38 39 40 Books 41 ----- 42 43 * **Outlier Detection in Python** by Brett Kennedy (Manning, 2024): chapters 6, 7, and 14 on PyOD. 44 * **Handbook of Anomaly Detection** by Chris Kuo (Columbia University): entire book built around PyOD. 45 * **Finding Ghosts in Your Data** by Kevin Feasel (Apress / O'Reilly): chapter 12 on COPOD. 46 * **Advanced Techniques for Anomaly Detection: Beyond the Basics** (Routledge / CRC Press, 2025). 47 * **Anomaly Detection: Recent Advances, AI and ML Perspectives** (IntechOpen, 2024). 48 49 50 Courses 51 ------- 52 53 * **DataCamp** -- `Anomaly Detection in Python <https://www.datacamp.com/courses/anomaly-detection-in-python>`_: dedicated PyOD chapter; DataCamp's platform reports 19M+ learners. 54 * **Manning liveProject** -- `Using PyOD and Ensembles Methods <https://www.manning.com/liveproject/using-pyod-and-ensembles-methods>`_: hands-on project. 55 * **O'Reilly Video Edition** -- Outlier Detection in Python, dedicated PyOD chapters. 56 * **Udemy** -- multiple courses including *Anomaly Detection: ML, DL, AutoML* and *Certified Anomaly Detection & Outlier Analytics*. 57 58 59 Podcasts & Talks 60 ---------------- 61 62 * **Talk Python To Me #497** -- `Outlier Detection with Python <https://talkpython.fm/episodes/show/497/outlier-detection-with-python>`_. 63 * **Real Python Podcast #208** -- `Detecting Outliers and Visualizing With PyOD <https://realpython.com/podcasts/rpp/208/>`_. 64 65 66 Media Coverage 67 -------------- 68 69 Articles and tutorials published by independent outlets: 70 71 * **Analytics Vidhya** -- `An Awesome Tutorial to Learn Outlier Detection in Python using PyOD <https://www.analyticsvidhya.com/blog/2019/02/outlier-detection-python-pyod/>`_ 72 * **KDnuggets** -- `An Overview of Outlier Detection Methods from PyOD <https://www.kdnuggets.com/2019/06/overview-outlier-detection-methods-pyod.html>`_ 73 * **KDnuggets** -- `Outlier Detection Methods Cheat Sheet <https://www.kdnuggets.com/2019/02/outlier-detection-methods-cheat-sheet.html>`_ 74 * **Towards Data Science** -- `Introducing Anomaly Detection in Python with PyOD <https://towardsdatascience.com/introducing-anomaly-outlier-detection-in-python-with-pyod-40afcccee9ff/>`_ 75 * **Towards Data Science** -- `Real-Time Anomaly Detection With Python <https://towardsdatascience.com/real-time-anomaly-detection-with-python-36e3455e84e2/>`_ (March 2025, PyOD + PySAD) 76 * **Towards Data Science** -- `Boosting Your Anomaly Detection With LLMs <https://towardsdatascience.com/boosting-your-anomaly-detection-with-llms/>`_ (September 2025, dedicated to PyOD 2's LLM-powered model selection) 77 * **Ericsson Blog** -- `How to make anomaly detection more accessible <https://www.ericsson.com/en/blog/2020/7/how-to-make-anomaly-detection-more-accessible>`_ 78 * **Elder Research** -- `Business Insights Meet Analytics Skills in Anomaly Detection <https://www.elderresearch.com/blog/business-insights-meet-analytics-skills-in-anomaly-detection/>`_ 79 * **Data Reply IT (Reply Group)** -- `Anomaly Detection made easy with PyOD <https://medium.com/data-reply-it-datatech/anomaly-detection-made-easy-with-pyod-960faf6da4e5>`_ 80 * **Cake.ai** -- `Anomaly Detection Software: A Complete Guide <https://www.cake.ai/blog/open-source-anomaly-detection-tools>`_ 81 * **Number Analytics** -- `Advanced Nonparametric Outlier Identification <https://www.numberanalytics.com/blog/advanced-nonparametric-outlier-identification>`_ (2025) 82 * **The Data Scientist** -- `Anomaly detection in Python using the PyOD library <https://thedatascientist.com/anomaly-detection-in-python-using-the-pyod-library/>`_ 83 * **SmartDev** -- `Master AI Anomaly Detection: The Definitive Guide <https://smartdev.com/ai-anomaly-detection/>`_ 84 * **Milvus / Zilliz** -- `Open-source libraries for anomaly detection <https://milvus.io/ai-quick-reference/what-are-opensource-libraries-for-anomaly-detection>`_ 85 86 87 International Reach 88 ------------------- 89 90 Beyond English, PyOD tutorials and translations exist in at least five non-English languages: 91 92 * **Chinese** -- 10+ tutorials across CSDN, Zhihu, 搜狐 (Sohu), 机器之心 (Jiqizhixin), 腾讯云开发者社区 (Tencent Cloud Developer), 智东西 (Zhidx), Bilibili. The community project `aidoczh.com <https://www.aidoczh.com>`_ maintains a full Chinese translation of PyOD documentation. 93 * **Japanese** -- 4+ tutorials including Qiita, Codemajin, DataPowerNow, Scutum, TRYETING, and ClassCat. 94 * **Korean** -- 3 tutorials (Tistory, JunPyoPark, DataNetworkAnalysis). 95 * **German** -- 5 sources including Hahn-Schickard / EmbedML, Konfuzio, Acervo Lima, KI Blog. 96 * **Spanish** -- Aprende Machine Learning and Medium tutorials. 97 98 99 Academic Follow-on Work 100 ----------------------- 101 102 * `Text-ADBench <https://arxiv.org/abs/2507.12295>`_ (Jicong Fan et al., July 2025): external follow-on benchmark inspired by ADBench. 103 * COPOD and ECOD cited as "most efficacious" methods in digital forensics research (*CEUR-WS Vol-4092*). 104 * Two-phase Dual COPOD Method for ICS security (arXiv:2305.00982). 105 * Graph Diffusion Models for Anomaly Detection (Amazon Science, 2024): cites BOND and PyGOD. 106 107 108 Platforms 109 --------- 110 111 * `Kaggle <https://www.kaggle.com/search?q=pyod>`_: 7+ dedicated public notebooks. 112 * `HelloGitHub <https://hellogithub.com>`_: featured open-source project. 113 114 115 Summary 116 ------- 117 118 As of April 2026: **38+ million downloads** on `PyPI <https://pepy.tech/project/pyod>`_, **9K+ stars** on `GitHub <https://github.com/yzhao062/pyod>`_, one *Nature Scientific Data* citation (ESA OPS-SAT), 3+ dedicated books, 2 major podcasts, 4+ online courses, tutorials in 5 non-English languages, and 60+ third-party media articles. PyOD has also been cited in research from USC Viterbi, Amazon Science, Microsoft Research, and Lawrence Livermore National Laboratory.