Machine Learning for Time Series Data Analysis—Best Practices in Prediction and Anomaly Detection Using Python
Published byO'Reilly Media, Inc.
What is this learning path about, and why is it important?
With organizations of all kinds looking to extract more value from the enormous volumes of data that they’re continually collecting, data analysis and machine learning have become hot areas of interest lately. Applying these paradigms to big data is helping companies across the span of industries to learn more about their products and customers. Time–series analysis is one of the classic topics in data analysis and machine learning that has huge business value. Classic methods relied on linear models, but real-world data is much more complex. In addition, recent advances in the field of deep learning have proven to work well in complex cases.
This learning path provides an overview of the current state-of-the-art of time–series data analysis. Entry- to intermediate-level practitioners can gain a solid understanding of the best use cases, the current toolsets, and best practices for handling time–series data. In this learning path, you will focus on growing libraries in Python, which is quickly becoming the de facto standard for practical data science with which you can quickly move models from prototype to production.
What you’ll learn—and how you can apply it
- The basics of time–series analysis, including the relevant types of data analysis problems, as well as classic approaches using autoregressive models
- Best practices and recommendations for time–series prediction and anomaly detection
- Recent advances in time–series analysis; in particular, structured inference and deep learning–based approaches
- How to utilize the relevant Python data analysis libraries and toolsets
- Key concepts related to time–series storage and databases
- Some relevant use cases from industries in order to gain a better understanding of the business value of analyzing time series data
This learning path is for you because…
- You are a data scientist, data engineer, or technical lead, involved with analyzing or processing time–series data, and you want an introduction to the topic, with hands-on recommendations for moving forward and the right toolsets and libraries
- You have experience with Python and want to expand your skills and learn how to process time–series data as it relates to the fast-growing arena of data analysis and machine learning
- You have some experience developing machine learning models, but would like to learn the current best practices to refine your models
- You should have basic familiarity with Python
- You should have beginner-level background in data analysis and data science
Materials or downloads needed in advance: None