XGBoost is an open-source library that makes predictions by adding up the answers of many small decision trees, trained one after another with gradient boosting.
In plain words
A single decision tree is a simple flowchart of questions. XGBoost builds many trees in sequence. Each new tree is added to fix what the trees before it got wrong, while the earlier trees stay fixed. To predict, it sums the scores from every tree. The same code can also run across a cluster for very large datasets.
A quick example
Imagine predicting whether a customer will cancel a subscription from a table of columns such as age, plan and months subscribed. The first tree makes a rough guess. The next tree focuses on the mistakes left over, and so on for hundreds of trees. The final prediction is the combined score of all of them.
Where you'll meet it
- Machine learning competitions and data science work.
- Large jobs run on Spark, Dask or Kubernetes clusters.
Related
Sources
- DocsXGBoost Documentation, XGBoost developers · read 29 Sept 2026
- DocsIntroduction to Boosted Trees, XGBoost developers · read 29 Sept 2026
- PaperXGBoost: A Scalable Tree Boosting System, Chen and Guestrin (arXiv) · read 29 Sept 2026
- Docsdmlc/xgboost, DMLC (GitHub) · read 29 Sept 2026
Checked by WhatIsAI against the sources above · 29 Sept 2026. Spotted an error? Report it.