Feature engineering
Feature engineering is the creation, transformation, or selection of input features from data to provide useful representations for analysis or modeling.
Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI
Feature processing is the transformation or elaboration of input features to make them suitable or more informative for machine learning models.
- How it is used
- Prepares tabular inputs for predictive modeling by extracting relevant information and aligning representations with model needs.
- How deep it goes
- Develops an end-to-end workflow from text extraction and aggregation through geographic features, while also naming advanced imputation and encoding methods.
- Examples
- Demonstrates Tokyo Airbnb price prediction with regex fields, keyword flags, review recency, price aggregation, geographic bins, and proximity distances.
- What you can do after
- Apply it via pandas, regular expressions, aggregation, and scikit-learn KDTree to create and assess tabular features for a prediction task.
Deep Learning with R, Third Edition
Feature engineering is the manual design or transformation of input features into representations suitable for machine learning.
- How it is used
- Uses feature engineering to simplify inputs and help an initial model beat a baseline.
- How deep it goes
- Defines hardcoded transformations and explains a progression from pixels to coordinates to angles, without implementing the clock transformation.
- Examples
- Illustrates clock reading through hand-tip coordinates and polar angles.
- What you can do after
- Evaluate/understand how problem knowledge can simplify inputs and guide feature choices for an initial model.
Read how Deep Learning with R, Third Edition explains feature engineering →
Deep Learning with Python, Third Edition
The manual design or transformation of input variables into representations better suited for a learning algorithm.
- How it is used
- Uses feature engineering to make model inputs easier to learn from and help an initial model beat a baseline.
- How deep it goes
- Explains non-learned transformations through a clock example and gives feature-selection guidance without implementing the clock script.
- Examples
- Uses clock-hand coordinates and angles to simplify time reading, ultimately replacing the model with rounding and a lookup.
- What you can do after
- Evaluate and understand how problem knowledge can suggest simpler input representations and when feature selection may help.
Read how Deep Learning with Python, Third Edition explains feature engineering →
Knowledge Graphs and LLMs in Action
Manual feature extraction is the hand-designed selection or computation of features from raw data.
- How it is used
- Converts nodes, relationships, or whole graphs into vectors for downstream classification, link prediction, and other machine-learning tasks.
- How deep it goes
- Descends from manual node metrics to homophily vectors, relationship combinations, metapaths, and DWPC, while contrasting semiautomated and neural alternatives.
- Examples
- Uses karate-club degree features, homophily-based vectors, Node2Vec comparisons, and Hetionet compound–disease metapaths with DWPC.
- What you can do after
- Apply it via NetworkX, tabular classifiers, and graph queries to build interpretable vectors and diagnose representation-driven performance.
Read how Knowledge Graphs and LLMs in Action explains feature engineering →
Natural Language Processing in Action, Second Edition
Feature engineering is the deliberate construction or transformation of input representations to make them useful for predictive modeling.
- How it is used
- Uses feature engineering to prepare representations for NLP pipelines and predictive models.
- How deep it goes
- Explains residual-guided transformations, polynomial interactions, and a hardcoded convolution, while contrasting manual features with learned extractors.
- Examples
- Illustrates squared features, polynomial interactions, and a two-point moving average over adverb indicators.
- What you can do after
- Apply feature transformations and a hardcoded convolution to explore patterns useful for modeling.
Read how Natural Language Processing in Action, Second Edition explains feature engineering →
Outlier Detection in Python
Feature engineering is the process of modifying or combining original data attributes to create inputs that are more useful for analysis or modeling.
- How it is used
- Creates detector-ready inputs that expose unusual values, relationships, frequencies, and temporal patterns.
- How deep it goes
- Works through tabular, date, string, and time-series transformations with Python examples, while leaving feature choice dependent on judgment and evaluation.
- Examples
- Uses staff expenses, baseball ratios, phone numbers, date gaps, weekly patterns, and air-quality measurements.
- What you can do after
- Build it from scratch by deriving interpretable columns that help detectors test suspected anomalies in tabular or time-series data.
Read how Outlier Detection in Python explains feature engineering →
Machine Learning System Design
Feature engineering is the process of constructing, transforming, or selecting input features for a machine learning model.
- How it is used
- Guides the selection and transformation of model inputs as part of ML system design.
- How deep it goes
- Defines an iterative process and gives planning, selection, validation, and maintenance considerations without detailed algorithm tutorials.
- Examples
- Uses weather forecasts for demand prediction and model-generated representations of users or items as examples.
- What you can do after
- Apply feature engineering by prioritizing iterations, checking data availability, and monitoring feature reliability.
Read how Machine Learning System Design explains feature engineering →
Fight Fraud with Machine Learning
Feature engineering is the process of manually extracting or constructing input variables from raw data for a machine-learning system.
- How it is used
- Supports classical fraud-detection models by converting raw records into explicit variables for training and prediction.
- How deep it goes
- Explains manual extraction, dataset augmentation, correlation checks, and feature-importance interpretation, while deferring detailed implementation code.
- Examples
- Uses phishing-email attributes, engineered fraud variables such as amountOrig and errorBalanceDest, and random-forest importance plots.
- What you can do after
- Builds explicit tabular inputs, checks their redundancy, and interprets their contribution to a trained fraud model.
Read how Fight Fraud with Machine Learning explains feature engineering →
Effective Data Analysis
Feature engineering is the creation or transformation of variables from existing data to better represent processes relevant to a model.
- How it is used
- Supports exploratory analysis and iterative model refinement by making organizational processes measurable at a useful analytical grain.
- How deep it goes
- Explains aggregation, extraction, encoding, and numerical transformation with examples, while stopping short of detailed algorithmic internals.
- Examples
- Uses rat-sighting counts, weather variables, date components, weekday indicators, dummy coding, boxplots, correlations, and wind-speed transformations.
- What you can do after
- Build it from scratch by deriving, transforming, encoding, and combining variables for explanatory or predictive modeling.
Read how Effective Data Analysis explains feature engineering →
Vibe Engineering
A feature engineering pipeline is a sequence of transformations that creates model inputs from raw data.
- How it is used
- Creates model inputs from raw medicine data as part of an LLM-assisted data science workflow.
- How deep it goes
- Works through field extraction, diagnostic counts, sample inspection, and saving processed data, while emphasizing verification of LLM-generated parsing.
- Examples
- Uses medicine composition strings to extract salt names and dosages and identify multi-salt products.
- What you can do after
- Apply it via Python and pandas to create and inspect features from medicine records.
Build AI Drug Discovery Pipelines
Feature engineering is the manual design or selection of input representations intended to expose task-relevant patterns to a predictive model.
- How it is used
- Supports predictive modeling for molecular properties, druglikeness, toxicity, solubility, and drug-target affinity.
- How deep it goes
- Explains feature choices and graph encoding through implementation details, while deferring exhaustive chemistry rules and learned-feature design.
- Examples
- Uses molecular weight, atom counts, rotatable bonds, substructures, and 78-dimensional atom vectors as concrete representations.
- What you can do after
- Build it from scratch by converting SMILES into molecular graphs and constructing normalized atomic feature vectors.
Read how Build AI Drug Discovery Pipelines explains feature engineering →
Financial AI in Practice
Feature engineering is the process of constructing informative input variables from raw data for statistical or machine-learning models.
- How it is used
- Uses feature engineering in financial data pipelines to turn raw information into signals for market prediction and fraud detection.
- How deep it goes
- Explains feature categories, signal rationale, and real-time ingestion, while illustrating market predictors and operational risks.
- Examples
- Illustrates failed payment counts, shared-device links, a bond volatility spread, and retrieval-ready news vectors.
- What you can do after
- Evaluate/understand candidate financial signals using domain meaning, timing, volume, correlation stability, and privacy considerations.
Read how Financial AI in Practice explains feature engineering →
Delivering Machine Learning Projects: From design to deployment
Feature engineering is the transformation of raw data into informative input variables that represent useful structure for a machine learning model.
- How it is used
- Uses feature engineering as a core ML-pipeline task that prepares raw data before algorithms consume it.
- How deep it goes
- Explains planning, pipeline implementation, bias handling, structural transformations, and augmentation, without deep algorithmic exposition.
- Examples
- Uses calendar sequences, directional values, and temperature plots to illustrate informative representations of cyclical or structured data.
- What you can do after
- Build it from scratch by planning, implementing, reviewing, and augmenting a feature-engineering pipeline.
Software Engineering for Data Scientists
Feature engineering is the creation or transformation of input variables to make them suitable or informative for machine learning.
- How it is used
- Prepares existing variables for model training by creating inputs that improve or enable prediction.
- How deep it goes
- Defines feature engineering and demonstrates date parsing, age bucketing, and categorical columns without detailing transformation algorithms.
- Examples
- Uses transaction dates for credit fraud, bucketed customer ages for churn, and site-category indicator columns for ad-click data.
- What you can do after
- Apply it via Dask or a data pipeline to create columns and transformations before training a model.
Read how Software Engineering for Data Scientists explains feature engineering →
Inside Deep Learning: Math, Algorithms, Models
The manual design of input transformations or derived features intended to improve model performance.
- How it is used
- Improves image-model learning through augmented training data and uses a Faster R-CNN backbone to provide features for downstream subnetworks.
- How deep it goes
- Demonstrates image-transform pipelines and validation gains, but does not develop a general feature-design procedure.
- Examples
- Uses rotations, translations, shears, crops, flips, perspective changes, color jitter, and small affine transforms on MNIST images.
- What you can do after
- Apply it via torchvision transforms to augment MNIST training images while keeping test transforms deterministic.
Read how Inside Deep Learning: Math, Algorithms, Models explains feature engineering →
Transfer Learning for Natural Language Processing
Feature engineering is the process of transforming input data into features suitable for machine-learning methods.
- How it is used
- Prepares inputs for shallow learning methods in NLP prediction tasks.
- How deep it goes
- Defines manual feature engineering and contrasts it with automatic feature learning, without specifying extraction steps.
- Examples
- Uses sentence classification and positive-or-negative sentiment analysis as NLP task contexts.
- What you can do after
- Evaluate manual feature engineering’s tradeoffs against automatically learned features.
Read how Transfer Learning for Natural Language Processing explains feature engineering →
Sutskever's List
Feature engineering is the process of designing input representations from raw data for use by a learning algorithm.
- How it is used
- Frames handcrafted representations as the input pipeline used by pre-AlexNet vision systems and as a contrast to learned representations.
- How deep it goes
- Explains descriptors, global image representations, classifier inputs, and a failure case, then contrasts them with learned features.
- Examples
- Uses SIFT and HOG examples, a water patch mistaken for a car, and AlexNet’s error-rate comparison with conventional methods.
- What you can do after
- Evaluate how a hand-designed representation can encode useful assumptions yet cause classification errors when its features mislead.
The Art of AI Product Development
Feature engineering is the process of creating, selecting, transforming, and refining input features to improve a machine-learning model's results.
- How it is used
- Improves clustering by turning product behavior and profile data into signals aligned with meaningful customer segments.
- How deep it goes
- Explains selection, enrichment, cleaning, scaling, and iterative assessment, without descending into algorithm implementation.
- Examples
- Uses recommendation reliance, visits, purchases, checkout activity, profile edits, order value, browsing behavior, and product categories.
- What you can do after
- Diagnose/debug weak clusters by removing irrelevant signals, correcting data problems, scaling features, and iterating with domain knowledge.
Read how The Art of AI Product Development explains feature engineering →
Applied Reinforcement Learning
Feature engineering is the construction or transformation of input features into representations suitable for a model or learning algorithm.
- How it is used
- Transforms gas station inventory and pending orders into a state representation for tabular learning.
- How deep it goes
- Works through an implementation that aggregates pending orders into total quantity and next arrival time.
- Examples
- Uses a gas station state with current inventory, pending fuel, and delivery arrival time.
- What you can do after
- Apply feature engineering to compact a tabular learner’s state while retaining information needed for decisions.
Read how Applied Reinforcement Learning explains feature engineering →
Distributed Machine Learning Patterns
Feature engineering is the process of creating, transforming, or selecting input features to improve a machine learning model.
- How it is used
- Improves model inputs by combining overlapping audiovisual features before fitting, while recognizing that deep learning may learn such relationships itself.
- How deep it goes
- Explains derived-feature construction and its performance rationale, then stops short of a procedural feature-engineering workflow.
- Examples
- Uses overlapping Animal and Pet audiovisual features and a 10-class LeNet classification scenario.
- What you can do after
- Evaluate whether combining overlapping audiovisual features is worthwhile before training, especially when exploration threatens single-GPU training time.
Read how Distributed Machine Learning Patterns explains feature engineering →
Digital Twins in Action
Feature engineering is the process of constructing informative input variables from raw data for analysis or machine learning.
- How it is used
- Transforms operational sensor data into informative inputs for digital-twin models.
- How deep it goes
- Works through temporal encoding, temperature lags, and a derived physical difference with Python listings, then notes automated alternatives.
- Examples
- Uses hourly home temperatures, cyclical hour values, and indoor–outdoor temperature differences to construct features.
- What you can do after
- Apply it using NumPy and pandas to encode time, create temperature lags, and calculate a temperature delta.
Read how Digital Twins in Action explains feature engineering →
Timeless Algorithms, The Foundational Ideas
Feature engineering is the process of transforming raw data into input features suitable for statistical or machine-learning models.
- How it is used
- Transforms collected observations into variables that expose patterns for learning methods.
- How deep it goes
- Defines preparation choices through churn and lending examples, but gives no implementation procedure or mathematical formalization.
- Examples
- Uses churn measures such as average spending and lending measures such as debt-to-income ratio as examples.
- What you can do after
- Build it from scratch by cleaning, encoding, filtering, and combining observations into task-relevant variables.
Read how Timeless Algorithms, The Foundational Ideas explains feature engineering →
Evolutionary Deep Learning: Genetic algorithms and neural networks
Feature engineering is the construction, transformation, or selection of input variables that represent useful information for a machine-learning model.
- How it is used
- Deploys feature construction and selection within automated machine-learning workflows to improve data preparation and model development.
- How deep it goes
- Defines expert-driven feature extraction and its automation, but defers implementation details to later treatment.
- Examples
- Uses expert selection of relevant features and a workflow figure showing feature engineering among AutoML and evolutionary deep-learning steps.
- What you can do after
- Apply feature engineering within AutoML and deep-learning optimization workflows.
Demand Forecasting Best Practices
Feature engineering is the selection, creation, or modification of input features for a machine learning model.
- How it is used
- Improves demand forecasts by supplying models with business-relevant information.
- How deep it goes
- Defines selection, creation, and modification, then gives brainstorming prompts but defers model-parameter tuning.
- Examples
- Uses Black Friday sale status, price changes, average demand, promotions, stockouts, and a typical-inputs table.
- What you can do after
- Apply it by collaboratively brainstorming and collecting business-relevant features for a demand forecast.
Read how Demand Forecasting Best Practices explains feature engineering →
Also covered in
- Deep Learning with PyTorch, Second Edition
- Cyber Threat Hunting
- Build AI-Enhanced Web Apps
- Data Without Labels
- How Large Language Models Work
- Grokking Algorithms, Second Edition
- Math and Architectures of Deep Learning
- Inside AI
- Learn AI-Assisted Python Programming, Second Edition
- Grokking Machine Learning
- The AI Pocket Book
- AI-Powered Search
What to know first
- language modelA language model is a computational model that represents patterns in linguistic sequences to assign probabilities, predict, generate, or evaluate language.
- machine learningMachine learning is a field of computational methods in which systems learn patterns or relationships from data to make predictions, decisions, or other outputs.
- inputAn input is information, a value, or an action supplied to a system, program, or computation for processing or to influence its behavior.
- computer visionComputer vision is the field of computing concerned with extracting and interpreting information from images, video, and other visual data.
Related concepts
- foundation modeluses · A foundation model is a broadly trained machine-learning model whose learned capabilities can be reused or adapted for a range of downstream tasks.
- feature selectionhas part · Feature selection is the process of choosing a subset of input features for use in a modeling, analysis, or prediction task.
- data layerpart of · A software architectural layer that mediates access to data by storing, retrieving, transforming, or exposing it to other components.
- language modeladdressed by · A language model is a computational model that represents patterns in linguistic sequences to assign probabilities, predict, generate, or evaluate language.
- signalproduces · A signal is an observable value, pattern, event, or attribute that conveys information about a condition, behavior, or phenomenon.
- pipelinepart of · A pipeline is an ordered sequence of connected processing stages through which data or work flows, with each stage passing or transforming its output for subsequent stages.
Questions about feature engineering
What is feature engineering?
Feature engineering is the creation, transformation, or selection of input features from data to provide useful representations for analysis or modeling.
Which Manning books explain feature engineering?
Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI, Deep Learning with R, Third Edition, Deep Learning with Python, Third Edition, Knowledge Graphs and LLMs in Action, Natural Language Processing in Action, Second Edition, Outlier Detection in Python, Machine Learning System Design, Fight Fraud with Machine Learning and 16 more.
What should I know before learning feature engineering?
Manning books that teach feature engineering usually introduce language model, machine learning, input and computer vision first.
What is feature engineering part of?
Feature engineering is part of data layer. A software architectural layer that mediates access to data by storing, retrieving, transforming, or exposing it to other components.
What are the parts of feature engineering?
Feature engineering includes feature selection.
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.























