Time Series Anomaly Detection

This month, we have a talk from Linda Zhou on Time Series Anomaly Detection. There will also be a lightning talk from Jeff Fischer.

Video of Presentations
From YouTube

Lightning Talk: Simple Automation using Argparse and Make

Speaker: Jeff Fischer

See you how can use Python's argparse module along with the ubiquitous Make utility to add simple automation to your projects.

Main Talk: An Introduction to Time Series Anomaly Detection

The goal of time series anomaly detection is to identify unexpected data points or patterns in time-based sequences of values or events. This talk is for Python developers with a basic understanding of machine learning. I will cover common use cases for time series anomaly detection (e.g. observability, IoT failure prediction, fraud detection, log analysis), key terminology (e.g. outlier, change point, seasonality, drift, univariate, multivariate), statistical and ML algorithms (e.g. ARIMA, LOF, Isolation Forest), and Python libraries providing implementations of these models (e.g. Scikit-learn, PyCaret, Merlion, PYOD). At the end of the talk, you should have enough background to know where to start when approaching these types of problems.

Speaker Bio: Linda Zhou

Linda is a Director of Data Science Engineering at Cisco in the AppDynamics product line. She has worked in various roles in engineering, product management, alliances, and solutions marketing. She received an MBA from Carnegie Mellon University and a BS in Computer Science and Engineering from Jinan University.

Code of Conduct


Interactions online have less nuance than in-person interactions. Please be Open, Considerate and Respectful. Also, please refrain from discussing topics unrelated to the Python community or the technical content of the meeting.