A new, very good book on time series indexing

“Time Series Indexing: Implement iSAX in Python to index time series with confidence” by Mihalis Tsoukalos is unlike any other. Its focus on time series indexing, namely on iSAX and SAX, makes it a unique book on the subject. The explanations and examples in this book are clear and concise. Its coverage of a topic not often treated in depth in other books makes it particularly useful.

iSAX is a popular and effective technique for indexing and mining very large collections of time series data. You may want to use iSAX when you have very large collections of time series data that cannot fit in memory, and you need a compact and index-friendly representation that preserves the shape and similarity of the original data. iSAX can also handle variable-length queries and support adaptive indexing, parallelism, and distribution. However, iSAX may not be suitable for time series that have high variability or complex patterns, as it may lose some information during the discretization process.

iSAX is based on a symbolic representation of time series called SAX, which reduces the dimensionality and noise of the original data. iSAX allows for both exact and approximate similarity search, as well as data mining tasks such as clustering, classification, anomaly detection, and motif discovery. iSAX has many advantages over other time series indexes.

As it is explained in the book, besides implementing iSAX and the SAX representation as Python 3 packages, this book shows how to work with time series at the subsequence level and understand the information presented in academic research papers.

I strongly recommend this book. You can find the links below to purchase it at Amazon.

To buy it at US Amazon: https://amzn.to/45aCrt2
To buy it at Amazon in Germany: https://amzn.to/3RGgkYn