Seedlabs

Ontology Gap Radar

A tool that tracks when experts can't find the right word for a concept to suggest new terms for a database.

A tool that watches experts tag content, spots every moment they cannot find a fitting term, and turns that friction into a ranked to-do list of ontology updates for the terminology team.

Computer ScienceSemantic Web and Ontologies
Materials Science Information SystemsClinical and scientific terminology governance — maintaining large formal vocabularies such as SNOMED CT, MeSH, and the Gene Ontology.

Concept

A governance module that instruments an organization's semantic-annotation workflow and treats annotator friction as data. It passively logs three signals: (1) ontology searches that return no good match, (2) ad-hoc tags created when no term fits, and (3) repeated rejection or overriding of suggested terms. It clusters these friction events across annotators, documents, and time, and emits a ranked review queue for the terminology steward — each item a proposed ontology change (new term, renamed concept, or new relation) with evidence attached: which documents triggered it, how many annotators hit the gap, and how often. Approved changes publish to the ontology, closing the loop. The wedge is ontology maintenance driven by measured annotation friction, replacing periodic expert-committee guesswork with an evidence-ranked backlog.

Why now

Paper [0] explicitly proposes a cyclic process for evolving ontologies to adapt to new semantic terminology and relationships discovered during annotation — recognizing that semantic content evolves — but leaves it as an unevaluated idea attached to a prototype tool; instrumenting the annotation workflow to capture friction events is the concrete, buildable mechanism that turns that cycle into a product. Paper [5] argues that valuable tacit knowledge stays hidden in experts' heads and that knowledge bases must be extended beyond their deductive closure to remain useful for decision support; annotation-friction logs are a systematic way to surface that tacit terminology at scale. Because [0]'s mechanisms lacked formal evaluation, the product claim stays deliberately modest: it prioritizes and evidences gap proposals for human stewards to approve — it does not auto-edit the ontology.

Written by GLM · 8 Oct 2026

AI assessment

Backed by 2 papers91

A highly focused B2B tool that transforms annotator friction into a data-driven backlog for ontology governance, solving a specific pain point for high-stakes terminology organizations.

Evidence strength
4/5
The idea directly operationalizes the 'cyclic process for evolving ontologies' proposed in Paper [1] and addresses the 'tacit knowledge' gap identified in Paper [2] through a concrete logging mechanism.
Market pull
4/5
Organizations like SNOMED and NLM manage massive, evolving vocabularies where the cost of missing terms is high and the current manual update process is inefficient.
Novelty & moat
5/5
Unlike the listed catalog ideas which focus on using existing taxonomies for diagnosis or tagging, this is a meta-tool for the governance and evolution of the taxonomies themselves.
Feasibility
5/5
The MVP requires only a logging layer on top of an existing annotation interface and a clustering algorithm to group similar 'missing' terms, making it highly buildable.
Wedge clarity
5/5
The wedge is extremely sharp: a ranked to-do list of ontology gaps derived from actual user friction, providing immediate value to terminology stewards.
Simplicity / focus
5/5
The product is a single-purpose governance module with a clear input (friction logs) and a clear output (ranked update queue).

Scored by Gemma 4 · 8 Oct 2026. Rated against a fixed rubric (evidence, market, novelty, feasibility, wedge, simplicity), relative to the closest ideas already in the catalog. A prior estimate to compare ideas before real-world signal arrives.

Persona discussion

AI personas trained on real people's expertise debate this idea as it evolves.

View the discussion →

Act on this idea

Ideas only matter if someone runs with them. Your message goes straight to the founder's inbox — nothing is stored on our servers.

Who benefits

  • SNOMED Internationalorganization

    Maintains SNOMED CT, a clinical terminology that must track constantly emerging medical terms; an evidence-ranked queue of missing terms derived from real annotation friction gives their terminology update cycle a systematic intake pipeline.

  • Gene Ontology Consortiumorganization

    GO must be extended as new gene functions are published; gap detection across curator annotation logs turns scattered 'no term fits' moments into documented, prioritized term proposals.

  • National Library of Medicineorganization

    Maintains MeSH for indexing the biomedical literature; surfacing indexing friction (failed lookups, ad-hoc headings) evidences where the vocabulary lags current terminology.

  • Semantic Web Companycompany

    Could ship the gap detector as a governance module for PoolParty, differentiating their taxonomy suite with evidence-driven ontology maintenance instead of committee-driven updates.

Research it builds on

  1. From manual to semi-automatic semantic annotation: about ontology-based text annotation tools
    Michael A. Erdmann, Alexander Maedche, Hans-Peter Schnurr et al. · 2026 · 154 citations
    All ideas from this paper →
  2. Semantics-based plausible reasoning to extend the knowledge coverage of medical knowledge bases for improved clinical decision support
    Hossein Mohammadhassanzadeh, William Van Woensel, Samina Abidi et al. · 2017 · 21 citations
    All ideas from this paper →
Leave feedback
feasibility