Coming soon · Book Profile

Data Mining for Business Analytics: Concepts, Techniques, and Applications

A practical, hands-on guide that teaches the core concepts, techniques, and applications of data mining for business analytics using R, organized around a disciplined predictive-modeling process.

A profile of this book is on the way.

Get the book →

What it’s about

This book demystifies data mining for business students and practitioners by treating it as a structured, repeatable process rather than a black box of algorithms. Beginning with how to define a business problem and prepare data, it walks readers through the full predictive-analytics workflow: data exploration and visualization, dimension reduction, performance evaluation, and a comprehensive toolkit of supervised methods (multiple linear and logistic regression, k-nearest neighbors, naive Bayes, classification and regression trees, neural nets, discriminant analysis, ensembles) and unsupervised methods (association rules, collaborative filtering, cluster analysis). It also covers time-series forecasting and emerging data-analytics domains like social network and text mining. Throughout, the authors emphasize the central data-mining danger—overfitting—and the disciplined use of data partitioning (training/validation/test sets) and honest performance metrics to ensure that models generalize to new records. Rich with real business cases and R code, it equips the reader to build, evaluate, select, and deploy models that actually inform decisions.

The through-line

Who it’s for: A business student, analyst, or manager who wants to use data to make better, more accurate decisions and to build predictive models that work in practice.
The problem: They face large, messy datasets and a bewildering array of methods, and need to build models that reliably predict or classify new records. They feel overwhelmed by the complexity of algorithms and uncertain whether their models will actually work on new data or are just fooling themselves.
The plan: Define the business purpose and obtain the right data.
Explore, clean, and reduce the dimension of the data.
Partition data into training, validation, and test sets to guard against overfitting.
Choose and run appropriate supervised or unsupervised methods.
Evaluate performance honestly on holdout data using task-appropriate metrics.
The payoff: The reader builds models that generalize to new data and genuinely inform decisions. · They can choose, compare, and tune the right method for the problem and data at hand. · They reliably detect and avoid overfitting and measure performance with the right metrics.

See our guide

Related profiles we’ve built

Additional reading

Competing on Analytics: The New Science of Winning · Thomas H. Davenport and Jeanne G. Harris
The authors' previous book, which provides the strategic context by describing the earliest and most aggressive adopters of analytics. This book builds on it by providing a 'how-to' guide for all organizations.
Sources of Power: How People Make Decisions · Gary Klein
Discusses decision-making in high-pressure situations where there is no time for systematic data gathering, providing a contrast to the analytical approach and showing when intuition is necessary.
The Black Swan: The Impact of the Highly Improbable · Nassim Nicholas Taleb
Argues that statistical analysis is limited because it cannot predict rare, high-impact 'black swan' events, serving as a cautionary note on the limits of analytics.
Moneyball: The Art of Winning an Unfair Game · Michael Lewis
A popular case study of how the Oakland A's baseball team used an analytical approach to player selection to compete with richer teams, illustrating the power of competing on analytics.
Why Great Leaders Don't Take Yes for an Answer · Michael Roberto
Describes how to foster a culture of constructive conflict and debate in decision-making processes, which is essential for an analytical culture where assumptions are tested and merit triumphs over politics.
The Visual Display of Quantitative Information · Edward Tufte
A foundational work on how to create clear visual representations of data, a key skill for communicating analytical findings effectively.
Super Crunchers: Why Thinking-By-Numbers Is the New Way to Be Smart · Ian Ayres
The book discusses how statistical analyses are replacing human intuition and expert judgment in decision-making, a core theme related to the discussion of 'Moneyball' and the demise of the expert.
Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed · James Scott
Documents how governments' fetish for quantification and data can lead to misguided and harmful policies, providing a deep historical context for the book's warnings about the 'dictatorship of data'.
The War Managers · Douglas Kinnard
A survey of U.S. generals' views on the Vietnam War, revealing that the 'body count' metric was seen as a worthless and inflated measure of progress, illustrating the dangers of relying on flawed data.
Thinking, Fast and Slow · Daniel Kahneman
Explains the cognitive biases that lead humans to see illusory causal links, which the author's argue big data correlations can challenge and disprove.