“Codeless Time Series Analysis with KNIME is a perfect storm of codeless data science meets time series analysis meets one of the most popular analytics platforms available,” said Matthew Mayo, Data Scientist and Editor-in-Chief of KDNuggets, in his review of our most recent book.
KNIME Analytics Platform is free and open-source software that you can download to access, blend, analyze, and visualize their data, without any coding. Its low-code, no-code interface makes analytics accessible to anyone, offering an easy introduction for beginners, and an advanced data science set of tools for experienced users.
Maarit Widmann and Corey Weisinger from KNIME collaborated with Professor Daniele Tonini from the University of Bocconi, Italy to write this practical guide to implementing forecasting models for time series applications in KNIME.
That journey is worth sharing.
Why do we need this book?
Time series analysis is a veteran in the data science field. It has never been hyped nor forgotten. Instead, the techniques used for decades are still relevant. On top of that, new techniques have been developed further, such as LSTM-based forecasting, and new technologies have become available, smart homes for example. Thus, time series analysis has drawn more and more attention among data scientists and business analysts across different industries. Understanding the end-to-end process has become the key for productivity.
It's an in-demand skill, and one which the majority of data scientists have not taken the time to master.
Matthew Mayo. KDNuggets
Why we at KNIME decided to write this book was also because we wanted to highlight the time series functionalities of KNIME Analytics Platform, which Corey has been working on in the past few years. We wanted to show how the time series components together with KNIME Python Integration make advanced time series analysis accessible without coding or math background. You might know the time series components already if you have attended the Introduction to Time Series Analysis course or our time series workshops and webinars, or if you have read our blog posts about time series topics. The book collects and extends the materials from these resources.
How did it happen?
The idea to write this book came from two directions. Firstly, our colleagues Kathrin Melcher and Rosaria Silipo had been cooperating with Packt to write the Codeless Deep Learning with KNIME book, and Packt asked us to write a similar book about time series analysis. Secondly, the KNIME community has given us good feedback about the time series functionalities and asked for more educational materials and examples. So we knew it would be well received and worth writing!
However, we wouldn’t have started with such a challenge without the commitment of the expert in advanced business analytics and machine learning, Professor Daniele Tonini from the University of Bocconi, Italy. Once we had him on board, we knew that we could make it. Daniele would write the theoretical chapters and Maarit and Corey would write about the use cases.
What's inside?
The book contains 14 chapters in 3 sections. The first section contains the basics of time series analysis by Daniele and introduction to KNIME Analytics Platform by Maarit. After that, in the next two sections, it is just learning by doing. The second section introduces statistical and machine learning based methods to solve real-world problems, such as temperature forecasting by SARIMA models and energy demand prediction by LSTM. In this section, you can find especially Corey’s handprint. The last section introduces use cases by Corey and Maarit that integrate KNIME Analytics Platform with external tools such as Spark and H2O for big data analytics.
The book makes it easy to digest different types of time series analysis without having to worry about having to learn how to code.
Abdul, Amazon Reviewer
In a nutshell, this book approaches time series analysis from a practical point of view. The main outcome of the book is how to practice time series analysis.
Who are the authors?
The three authors of the book are also the instructors of the Introduction to Time Series Analysis at KNIME.
On top of that, Corey leads the KNIME Press team at KNIME and publishes regularly about technical topics, such as SARIMA, Fourier Transform, and IoT.
Maarit leads the Education team at KNIME and produces educational materials to effectively deliver the concepts around data science and KNIME Analytics Platform.
Daniele Tonini is one of the partners at Target Research, a boutique data science and analytics company based in Milan, and Contract Professor at Bocconi University (DEC Department) since 2008 and at University of Insubria since 2013.
We’re living in the golden era of data analytics, with plenty of data and algorithms of any kind... but topics like deep learning, artificial intelligence and NLP are attracting basically all of the attention of the practitioners, while the concept of Time Series forecasting is often neglected.
Daniele Tonini
In the last 15 years, he designed and deployed predictive analytics systems, data quality management and dynamic reporting tools, mainly for Customer Intelligence, Risk Management and Pricing applications. He currently teaches Statistics, Predictive Analytics for Data Driven Decision Making, Big Data and Databases, Market Research and Quantitative Methods for Management.
How was the writing experience?
It was a long, non-linearly progressing commitment! Although most of the use cases and knowledge was already there, it took several iterations to perfectionate the demonstrations and explanations. All for your best reading experience! You can read more about the writing experience in Packt's interviews with the authors, Corey, Maarit, and Daniele.