Technical & data
Embedding
FRReprésentation vectorielle d'un texte (souvent 1 024 à 3 072 dimensions) qui capture son sens. Les moteurs IA convertissent les contenus indexés et la question de l'utilisateur en embeddings pour mesurer leur similarité sémantique.
ENVector representation of a text (often 1,024 to 3,072 dimensions) capturing its meaning. AI engines convert indexed content and the user's question into embeddings to measure semantic similarity.
See also
More terms — Technical & data
- JSON-LDStructured data format recommended by Google and used by AI engines to understand page content (organization, product, FAQ, event). On ManuGEO, each profile publishes a Schema.org JSON-LD block describing the company, its products and certifications.
- robots.txtStandard file at a website's root that tells crawlers (Googlebot, Bingbot, and now ClaudeBot, GPTBot…) which URLs they may visit.
- Schema.orgShared vocabulary (backed by Google, Microsoft, Yahoo, Yandex) defining structured data types: Organization, Product, FAQPage, BreadcrumbList, LocalBusiness, etc. Foundation of JSON-LD.