Sayan Putatunda - Practical Machine Learning for Streaming Data with Python: Design, Develop, and Validate Online Learning Models

Any source code or other supplementary material referenced by the author in this book is available to readers on GitHub via the books product page, located at www.apress.com/978-1-4842-6866-7 . For more detailed information, please visit http://www.apress.com/source-code .

Advancements in technology have made the continuous collection of data possible. We are inundated with data in various daily transactions (e.g., POS transactions in retail outlets like Walmart, Target, etc.), sensor data, web data, social media data, stock prices, search queries, clickstream data, and so forth. This is all a source for high-velocity data; that is, streaming data.

This book is a quick-start guide for beginners to learn, understand, and implement various machine learning models on streaming data to generate real-time insights using Python. Streaming data is defined and infinite and continuous inflow of data at a high pace. Most of the machine learning models that we deal with are in the batch context. However, batch learning models are not suitable for handling streaming data as multiple passes over the data are not possible. The batch models may soon become outdated due to concept drift (i.e., the data distribution changes over time). Other challenges of streaming data include memory limitations and the high speed of data accumulation, which led to the development of a different class of methods, known as incremental or online learning methods.

The industrys adoption of various online learning/incremental learning methods has been quite slow over the years, but things are now changing quickly due to the vast applications of real-time machine learning. So, these techniques should be a part of a data scientists repertoire!

This book is suitable for data scientists, machine learning engineers, researchers, software engineers, academicians, and data science aspirants with basic programming skills in Python and keen on exploring machine learning with streaming data for a career move or an enterprise/academic project.

This book focuses on an end-to-end accelerated track delving into a holistic approach to develop concept drift detection algorithms, supervised learning algorithms (for regression and classification tasks), and unsupervised learning algorithms implemented in Python. Overall, this book comprises four chapters.

Chapter introduces the concept of streaming data, its various challenges, some of its real-world business applications, and various windowing techniques. This chapter also introduces the concepts of incremental and online learning algorithms and the scikit-multiflow framework in Python.

Chapter covers the various change detection/concept drift detection algorithms and their implementation on various datasets using scikit-multiflow.

Chapter discusses the various regression and classification algorithms for streaming data (including ensemble learning) and its implementation on various datasets.

Finally, Chapter introduces unsupervised learning for streaming data. It also provides a brief overview of other open source tools for handling streaming data, such as Spark Streaming, Massive Online Analysis (MOA), and more.

I would like to thank all those who have contributed to bringing this book to publication for their help, support, and input. I would like to give a heartfelt dedication to my mother (Mrs. Mili Putatunda) and my father (Mr. Supriya Putatunda) for their lifelong love, support, and encouragement in all aspects of my life. My special thanks to my wife (Mrs. Srijani Putatunda) for her support and encouragement. I am indebted to my thesis adviser, Prof. Arnab Kumar Laha, for introducing me to this wonderful research topic (i.e., analysis of streaming data) during my PhD days at IIM Ahmedabad and for having countless captivating and patient conversations with me.

I would also like to thank and appreciate the efforts of the editors of this book, Celestin John, James Markham, and Shrikant Vishwakarma, at Apress/Springer for their essential encouragement.

Dr. Sayan Putatunda

Kolkata, India

January 2021