Data Science for Business
About this book
Written for the people who work with data scientists rather than for the people who will build the models. It teaches the fundamental concepts, including how a business problem maps onto a data mining task, why overfitting happens and how holdout evaluation guards against it, and how similarity, clustering and probability estimation actually work. The treatment of evaluation is unusually good, covering expected value calculations, lift and profit curves, and why accuracy is the wrong measure when classes are unbalanced. Worked examples run through churn prediction and targeted marketing from framing to business case. There is enough maths to follow the reasoning and not so much that it becomes a textbook.
Description via Product Digest.
Topics
Related concepts
- Supervised, unsupervised and reinforcement learningThe three ways a model learns, what each one needs from the data, and how to tell which problem is actually in hand.
- Measuring an AI featureAcceptance rate and the unit it is counted in, edit and override rate, the error rate a team has to build because nothing emits an event for a wrong answer, the arithmetic of the trade between latency and quality, and why ordinary engagement measures point the wrong way for a generative feature.
- Goal misgeneralisationThe failure where a system learns a goal that fits all of its training data and turns out to be the wrong goal once the distribution shifts.
- Evaluating a machine learning modelWhy accuracy is usually the wrong number, what precision and recall are really trading against each other, why the threshold is a product decision and the business metrics that decide whether any of it was worth doing.
- Transparency and explainabilityWhy an accurate model nobody can explain is sometimes the wrong choice, the difference between the two words, and what a model card is for.