{
  "patent_number": "US 6523026",
  "country": "US",
  "title": "How Computers Find Hidden Connections Between Different Fields of Knowledge",
  "original_title": "Method for retrieving semantically distant analogies",
  "summary": "A method for finding related ideas in completely different subjects by using math to map how words appear together, even when the subjects use different vocabulary.",
  "what_it_does": "This patent describes a way to find 'analogies' between two unrelated fields, such as finding a biological solution to a mechanical engineering problem. It works by first analyzing a source domain (like a library of biology papers) to create a high-dimensional map of how words appear together. It then uses this map—represented as vectors in a multi-dimensional space—to search a second, completely different domain (like a library of engineering patents) where the same words might not even appear. By comparing the 'meaning' of the words based on their context in the first domain, the system can pull up relevant documents from the second domain that share a functional relationship despite having no overlapping keywords.",
  "what_it_does_not_cover": [
    "Does not cover simple keyword-based search engines that rely on matching exact words or synonyms.",
    "Does not cover systems that require the target domain to contain the same vocabulary as the source domain.",
    "Does not cover manual categorization or human-led tagging of documents.",
    "Does not cover basic vector search that does not specifically map relationships across semantically distant domains."
  ],
  "filed": "2000-10-02",
  "granted": "2003-02-18",
  "expires": null,
  "status": "active",
  "holder": "Huntsman International LLC",
  "holder_url": "https://patentbrief.org/company/huntsman-international-llc",
  "inventors": [
    {
      "name": "Herbert R. Gillis",
      "url": "https://patentbrief.org/inventor/herbert-r-gillis"
    }
  ],
  "times_cited": 456,
  "tags": [
    "ai_ml",
    "software",
    "telecommunications"
  ],
  "abstract": "A process of identifying terms or sets of terms in target domains having functional relationships (roles) analogous to terms (contained in the query) selected from a source domain whereby queryrelevant but semantically distant (novel) analogies may be retrieved, corresponding to any user defined query. The process is capable of discovering deep functional analogies between terms in source and target domains, even where there is a misleading superficial matching of terms (i.e. same terms, with different meanings) between the query and the target domains. The process comprises the automated generation of abstract representations of source domain content, and application of the abstract representations of content to the efficient discovery of analogous objects in one or more semantically distant target domains. Said abstract representations of terms are preferably vectors in a high dimensionality space, encapsulating characteristic occurrence patterns of terms in the source domain.",
  "url": "https://patentbrief.org/patent/us/6523026/google-search-query-processing",
  "markdown_url": "https://patentbrief.org/patent/us/6523026/google-search-query-processing/md",
  "google_patents_url": "https://patents.google.com/patent/US6523026",
  "relatedPatents": []
}