Time series data are everywhere. Whether it is from sensors on automated vehicles and manufacturing equipment, meteorological data, or financial data from the equities market, it helps us understand the behavior of a system over time. However, real-world time series data can have many issues like missing data, outliers, noise, etc. The data needs to be cleaned and prepped first before it can be analyzed or used for model development. Unfortunately, it is not always clear how to clean this data. Which algorithm should be used for filling missing values? Should outliers be removed first or noise? How is data that is measured using different sample rates synchronized? The process is iterative and can be very time consuming. In this session, we will show you how to use timetables with the new Data Cleaner app and Live Editor tasks to identify and fix common issues in time series data. We will cover different data cleaning methods using both code and low-code techniques that can make the data prep process more efficient.
