Data Science Algorithms in a Week: Top 7 Algorithms for Scientific Computing, Data Analysis, and Machine Learning, Second Edition

Choosing an algorithm is part of solving a data problem—not a step to leave to guesswork. In Data Science Algorithms in a Week, Dávid Natingga uses practical examples and Python implementations to introduce seven widely used methods, from classification and clustering to regression and time-series analysis.

The second edition moves through the material with examples drawn from everyday and technical settings, then gives readers problems to work through. It is a useful grounding for readers who already know Python and have some familiarity with statistics, and want to understand how these algorithms work and when they may fit a task.

Seven approaches, one practical tour

The chapters progress through k-nearest neighbors, Naive Bayes, decision trees, random forests, k-means clustering, regression, and time-series analysis. Rather than treating the methods as names to memorize, the book develops them through examples such as classifying text, grouping household data, predicting values, and examining business trends.

See the decisions behind the algorithms

Worked examples bring attention to choices that affect analysis: selecting a value of k, rescaling data, choosing distance measures, handling inconsistent records, and interpreting trends or seasonality. The sequence helps connect the mechanics of an algorithm with the data problem it is meant to address.

Python practice, with room to test your understanding

Implementation sections and chapter problems give readers opportunities to follow the methods in code and then apply or examine the ideas themselves. Supporting appendices include a Python reference, statistics topics such as Bayesian inference and cross-validation, and a glossary of data-science algorithms and methods.

A grounded starting point for data-science study

This book is suited to aspiring data-science professionals who are comfortable with Python and have a little statistics background. It may also help working practitioners broaden their familiarity with common machine-learning techniques and the kinds of classification, clustering, prediction, and time-series questions those techniques address.