Traditional machine learning

Q: How deeply does Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI cover traditional machine learning?

Defines the approach and emphasizes feature engineering, then applies linear models and gradient boosting without lengthy mathematical detail.

As explained in Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI

Classical machine learning is the family of non-neural-network machine learning approaches that commonly rely on statistical optimization, similarity comparisons, and engineered features.

Traditional machine learning covers non-neural approaches for mapping tabular inputs to predictions. It includes methods based on statistical optimization or similarity comparisons, from linear regression to gradient-boosting decision trees. For business data stored in CSV files or relational tables, these methods can predict inventory needs, commodity prices, listing prices, and other operational targets. section 1.0 section 1.6

Feature engineering is central: transform raw columns into representations a model can use, retain informative signals, reduce noise, address nonlinear relationships, and incorporate domain knowledge. This work matters because model assumptions and data characteristics determine whether a representation is useful. section 1.3

Choose traditional machine learning when its trade-offs fit the problem:

Use linear models as a challenging baseline and as robust production candidates. section 4.0
Consider gradient boosting, including XGBoost, as a conventional choice for tabular data. section 1.6
Give it particular consideration for smaller datasets, where deep learning may struggle with limited rows and many parameters. section 12.1
Compare alternatives on simplicity, transparency, and efficacy rather than treating either classical or deep learning as universally best. section 3.0

Airbnb price prediction

A Tokyo Airbnb regression task has 10,000 rows and 31 columns, compared with 48,000 rows and 18 columns for an NYC classification task. More columns increase preprocessing and feature-engineering demands; fewer rows can favor classical approaches. An optimized XGBoost result substantially improved on a linear-regression baseline, reaching R-squared 0.729 versus 0.320. section 12.1 section 12.1

Table 12.1 Airbnb NYC problem vs Airbnb Tokyo problem (view table figure)

	Airbnb NYC	Airbnb Tokyo
Rows in dataset	48,000	10,000
Columns in dataset	18	31
Target	Classification: predict whether the price is over or under the median price	Regression: predict price

Compare the row counts, column counts, and prediction targets that shape the Airbnb modeling trade-off. section 12.1

Generative AI can assist with tabular workflows involving textual input or output, but it is not yet a general replacement for supervised classical models in tabular prediction. A cited benchmark reports stronger average F1 for XGBoost trained on all data than for few-shot ChatGPT or Flan-T5 on its prediction tasks. section 1.5

How Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI covers it

Cover of Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI

liveBook manning.com

Read Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI on liveBook Get the book at manning.com

Related concepts

deep learningcontrasts with · Deep learning is a machine-learning approach that uses neural networks with multiple computational layers to learn representations from data.
language modelcontrasts with · A language model is a computational model that represents patterns in linguistic tokens to estimate sequence probabilities and predict or generate language from context.
feature engineeringuses · Feature engineering is the creation, transformation, or selection of input features from data to provide useful representations for analysis or modeling.
structured datauses · Structured data is information organized according to an explicit schema or predictable fields, types, records, or relationships for consistent machine processing.
cross-validationused by · Cross-validation is a resampling technique that evaluates a model by repeatedly fitting it on some portions of available data and assessing it on held-out portions.
machine learningkind of · Machine learning is a field of computational methods in which systems learn patterns or relationships from data to make predictions, decisions, or other outputs.

This explainer was generated by Manning's knowledge graph from the text of Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI. The section links point to the parts of the book it draws on.