Encyclopedia · 176 concepts

AI Agents & Applications · intermediate · concept 154 of 176

AI-Powered Search & Ranking

Using ML to improve search results and ranking. Moved from keyword matching to semantic understanding. Encoder models moved search from keyword matching to semantic understanding, and vector search enables meaning-based retrieval. The frontier has moved again: major engines now generate an answer as the primary result and cite sources into it, so visibility increasingly means being cited rather than being ranked.

Key terms

Semantic searchLearning to rankBM25Cross-encoderBi-encoder

Courses, papers, and more