[1] Wrapper induction is the problem of devising extraction procedures on an automatic basis, with minimal reliance on hand-crafted rules.
Many web pages are automatically generated from structured data – telephone directories, product catalogs, etc.
– wrapped in a loosely structured presentation language (usually some variant of HTML), formatted for human browsing and navigation.
Structured data are typically descriptions of objects retrieved from underlying databases and displayed in web pages following fixed templates at a low level, injected into pages where the high-level structure can vary from week to week, per the rapidly evolving fashion of the site's presentation skin.
Due to these shortcomings, researchers have studied automated wrapper generation using unsupervised pattern mining.