Search

Semantic similarity search across all paragraphs

Changed on

Search the Urantia Papers using semantic similarity (vector embeddings). Returns conceptually related results even without exact keyword matches.

A reference such as 99:1.1 gives those paragraphs (match: "reference"). A pasted paragraph of 25 words or more gives itself first (match: "exact"). Each result has strength: strong, related, or weak. The list holds the 100 nearest paragraphs; meta.total is the length of that list, and meta.related counts those that are related or better.

Optional filters: paperId, partId.

post/search/semantic

Request

  • Base URL: https://api.urantia.dev
  • URL: https://api.urantia.dev/search/semantic
  • Auth: none required

Request body

qstring required
pageinteger
limitinteger
paperIdstring
partIdstring
includestring
lang'eng' | 'es'

Language of the results. es gives each result the Spanish text and titles; a question can be in Spanish or in English. strength and meta.related then allow for the lower similarity of a Spanish question.

Response

Semantic search results with pagination

Example response

{
  "data": [
    {
      "bibleParallels": [
        {
          "url": "https://ebible.org/eng-web/JHN17.htm#V1"
        }
      ],
      "scriptureParallels": [
        {
          "url": "https://en.wikisource.org/wiki/The_Sikh_Religion/Volume_1/Japji#:~:text=Make%20contentment%20and%20modesty%20thine"
        }
      ]
    }
  ]
}

Changes