Content understanding finds relevant content through keywords, language patterns, and known attribute matching. It is one dimension of user-intent understanding, alongside understanding users and domain-specific meaning. section 1.2.5
Search systems can build on content matching through a progression of capabilities:
- Start with keyword search, then tune field weights, boosts, and text or language analysis. section 1.3
- Add domain-specific resources such as synonym lists, taxonomies, known entities, and business rules. section 1.3
- Improve query interpretation and relevance with query classification, semantic parsing, user signals, testing, and machine-learned models. section 1.3
Distinguishing content from domain understanding
Matching a query against keywords uses content understanding. A structured query over known tags, categories, or entities relies on domain understanding. section 1.2.5
Figure 1.7 The dimensions of user intent: a combination of content understanding, user understanding, and domain understanding
The diagram places content understanding alongside user understanding and domain understanding as dimensions of user intent. section 1.2.5
