Logistic regression
Logistic regression is a statistical classification model that estimates categorical outcome probabilities by applying a logistic or softmax function to a linear combination of input features.
Grokking Machine Learning
A perceptron-like model that outputs a continuous value between 0 and 1 rather than a discrete class label.
- How it is used
- Uses continuous scores to classify labeled data while conveying more than a 0-or-1 prediction.
- How deep it goes
- Develops the sigmoid prediction, log loss, iterative weight updates, a Python implementation, and a library application.
- Examples
- Works through word-count sentiment data and IMDB reviews, then compares a Titanic survival model with other classifiers.
- What you can do after
- Build it from scratch with Python updates or apply it via Turi Create and Scikit-Learn, then check predictions and validation metrics.
Read how Grokking Machine Learning explains logistic regression →
Statistics Slam Dunk
Logistic regression is a statistical model that predicts the probability of a categorical outcome, typically by applying a logistic function to a linear predictor.
- How it is used
- Predicts binary outcomes such as postseason qualification while comparing offensive and defensive predictors.
- How deep it goes
- Explains binary targets, probabilities, maximum likelihood, log odds, fitting, classification, and evaluation metrics without deriving estimation procedures.
- Examples
- Uses NBA teams’ points allowed and points scored to predict postseason qualification and compare two simple models.
- What you can do after
- Apply it via base R glm() to fit binary classifiers, threshold probabilities, and evaluate predictions with confusion-matrix and ROC metrics.
Read how Statistics Slam Dunk explains logistic regression →
Grokking Statistics
A regression method that models the probability of a categorical outcome using a logistic link function.
- How it is used
- Classifies binary outcomes by estimating a label probability and applying a decision threshold.
- How deep it goes
- Explains binary data, sigmoid curves, thresholds, coefficients, plotting, and Python fitting; defers regularization details and multiclass classification.
- Examples
- Uses spam labels and a toddler-meltdown dataset relating hours without a nap to binary outcomes.
- What you can do after
- Apply it via scikit-learn to fit a binary classifier, inspect coefficients, plot its curve, and predict probabilities.
Natural Language Processing in Action, Second Edition
Logistic regression is a statistical classification model that estimates the probability of a class from a weighted combination of input features.
- How it is used
- Connects weighted features to probability-based classification, neural neurons, NLP pipelines, and simple decision gates.
- How deep it goes
- Explains sigmoid behavior, parameter effects, feature preparation, and model equivalence while deferring detailed training mechanics.
- Examples
- Uses clickbait headlines, fake-news classification, recorded name counts, character n-grams, and plotted logistic curves.
- What you can do after
- Apply it with Python feature vectors and scikit-learn, or build its sigmoid component from scratch and inspect parameter effects.
Read how Natural Language Processing in Action, Second Edition explains logistic regression →
Causal Inference for Data Science
Logistic regression is a statistical model that estimates the probability of a categorical, often binary, outcome using a logistic function of input variables.
- How it is used
- Estimates treatment probabilities from confounders to support propensity-score construction in healthcare and causal analysis.
- How deep it goes
- Explains binary probability prediction, model comparison, cross-validation, and AUC, while stopping short of detailed coefficient interpretation.
- Examples
- Uses a patient treatment table and synthetic variables where confounder Z influences binary treatment X and outcome Y.
- What you can do after
- Apply it to estimate each patient’s probability of receiving treatment and compare the resulting classifier with alternatives.
Read how Causal Inference for Data Science explains logistic regression →
Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI
Logistic regression is a regression model that estimates class probabilities using a logistic link function.
- How it is used
- Deploys logistic regression as a simple, effective baseline for tabular classification and a template for the standard modeling workflow.
- How deep it goes
- Explains probability output, link-function framing, preprocessing, evaluation, penalties, and solvers, but defers detailed optimization mechanics.
- Examples
- Demonstrates synthetic binary data, standardized features, encoded tabular columns, cross-validation scores, and alternative solver configurations.
- What you can do after
- Apply it via Scikit-learn to build, evaluate, and refit a processed tabular classification pipeline.
Effective Data Analysis
Logistic regression is a statistical model for estimating the relationship between a categorical outcome and one or more predictors.
- How it is used
- Uses predictors to estimate or predict a categorical outcome in an analytics or statistical modeling problem.
- How deep it goes
- Defines the model and its role, relates it to linear regression, and defers detailed regression methods.
- Examples
- Uses advertisement-click prediction from user age and viewing frequency as a categorical-outcome example.
- What you can do after
- Evaluate whether a categorical-outcome prediction problem calls for logistic regression rather than linear regression.
Read how Effective Data Analysis explains logistic regression →
Data Without Labels
A classification method that models class-event probabilities using a logistic or sigmoid transformation of input variables.
- How it is used
- Frames customer-churn prediction as binary supervised classification using historical customer attributes.
- How deep it goes
- Defines the sigmoid, logit, odds, intercept, and coefficient equations, but does not detail model fitting or evaluation procedures.
- Examples
- Uses a structured telecom dataset containing revenue, service duration, average cost, usage, and churn labels.
- What you can do after
- Evaluate customer churn probabilities from historical labeled attributes and distinguish bounded logistic outputs from linear predictions.
Ensemble Methods for Machine Learning
Logistic regression is a statistical classification model that estimates class probabilities by applying a logistic function to a linear combination of input features.
- How it is used
- Classifies binary outcomes by estimating class probabilities from feature values.
- How deep it goes
- Explains the logit, sigmoid, Bernoulli assumption, additive weights, and odds interpretation, then demonstrates library training.
- Examples
- Uses diabetes diagnosis, breast-cancer diagnosis, a one-feature probability figure, and a scikit-learn training listing.
- What you can do after
- Apply it via scikit-learn, estimate binary-class probabilities, and interpret feature effects through changes in odds.
Read how Ensemble Methods for Machine Learning explains logistic regression →
Starting Data Analytics with Generative AI and Python
Logistic regression is a regression method that models the probability of a categorical outcome, commonly a binary outcome.
- How it is used
- Uses categorical-outcome modeling to investigate e-commerce efficiency, completion rates, and customer satisfaction.
- How deep it goes
- Positions logistic regression as an inferential technique for identifying predictors, without explaining estimation, assumptions, or implementation.
- Examples
- Examines order completion, payment methods, delivery time, product price, and customer satisfaction in Brazilian e-commerce data.
- What you can do after
- Evaluate potential predictors of categorical business outcomes and connect findings to cautious recommendations.
Read how Starting Data Analytics with Generative AI and Python explains logistic regression →
Machine Learning Algorithms in Depth
A probabilistic classification model that maps inputs to class probabilities using a logistic or softmax transformation of a linear score.
- How it is used
- Uses probabilistic modeling to classify inputs into discrete class labels.
- How deep it goes
- Derives posterior probabilities, likelihood, gradients, and SGD updates before showing an implementation and visual results.
- Examples
- Demonstrates two Gaussian feature clusters, SGD training, cost reduction, and a learned decision boundary.
- What you can do after
- Build it from scratch with stochastic gradient descent and visualize its cost and decision boundary.
Read how Machine Learning Algorithms in Depth explains logistic regression →
The Art of AI Product Development
Logistic regression is a statistical classification model that estimates class probabilities using a logistic function of input features.
- How it is used
- Supports simple classification tasks where cost, privacy, speed, and interpretability influence model selection.
- How deep it goes
- Explains probability outputs, feature contributions, and trade-offs with neural networks, but stops short of mathematical or implementation detail.
- Examples
- Uses sentiment analysis, user segmentation, probability estimates, and feature-contribution examples.
- What you can do after
- Evaluate classification-model trade-offs and select an interpretable baseline when slightly lower accuracy is acceptable.
Read how The Art of AI Product Development explains logistic regression →
Related concepts
- maximum likelihood estimationuses · Maximum likelihood estimation is a statistical method that selects parameter values maximizing the likelihood of observed data under a specified probabilistic model.
- classifierkind of · A classifier is a predictive computational model that assigns inputs to one or more discrete classes or categories.
- sigmoiduses · A sigmoid is a smooth, bounded, S-shaped function that maps real-valued inputs between two limiting values, commonly zero and one.
- generalized linear modelkind of · A generalized linear model combines a linear predictor, a link function, and a response distribution from the exponential family.
- linear regressioncontrasts with · Linear regression is a statistical modeling method that estimates a continuous response as a linear combination of predictor variables and fitted coefficients.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.