<h1 id="using-bert-models-to-automatically-classify-domain-concepts-into-dolce-top-level-concepts-a-study-of-the-oaei-ontologies">Using Bert Models to Automatically Classify Domain Concepts into Dolce Top Level Concepts a Study of the Oaei Ontologies<a aria-hidden="true" class="anchor-heading icon-link" href="#using-bert-models-to-automatically-classify-domain-concepts-into-dolce-top-level-concepts-a-study-of-the-oaei-ontologies"></a></h1>
<ul>
<li><a href="https://ceur-ws.org/Vol-3637/paper32.pdf">https://ceur-ws.org/Vol-3637/paper32.pdf</a></li>
<li>keywords: <a href="/wiki/notes/3fa0re3cagbpf2ne3f0rsfe">Foundational Ontology</a></li>
</ul>
<h2 id="abstract">Abstract<a aria-hidden="true" class="anchor-heading icon-link" href="#abstract"></a></h2>
<p>Top-level ontologies provide a set of foundational concepts that have a well-founded philosophical meaning, being a useful tool in ontology engineering. However, in practice, few domain ontologies integrate top-level concepts. One of the difficulties refers to the selection of appropriate top-level concepts. This paper presents an analysis of top-level categories of a set of well-known domain ontologies from ontology matching benchmarks. Our main hypothesis is that training classification models using only concept comments (i.e., rdfs:comment) from top-level concepts can improve reported results in the literature. We then consider the best classifiers to estimate the distribution of concepts from Ontology Alignment Evaluation Initiative (OAEI) ontologies aligned to DOLCE top-level concepts.</p>
<hr>
<strong>Backlinks</strong>
<ul>
<li><a href="/wiki/notes/mjcpuwi7bne4r8moc0uruzx">DOLCE</a></li>
</ul>

Using Bert Models to Automatically Classify Domain Concepts into Dolce Top Level Concepts a Study of the Oaei Ontologies

public-notes


Welcome! And beware!  

## Purpose

This vault is the "[[learning-in-public|ar.swyx.learn-in-public]]" part of my Second Brain. If you share some of my interests, please [reach out](https://djradon.github.io)! 

## Contents

- [[t]] my personal wikipedia
- [[idea]] (not necessarily mine)
- Resource Notes, including highlights, thoughts and metadata, for various things:
  - [[ar]] mostly articles, papers and web pages
  - [[book]]
  - [[community]]
  - [[course]]
  - [[event]]
  - [[game]]
  - [[loc]]
  - [[org]]
  - [[user]] 
  - [[prdct]] mostly software products, but also ontologies
  - [[recipe]]
  - [[video]] movies, tv episodes, youtube, etc
- [[vs]]
- [[wanted]]
- Some "Metadata Notes":
  - [[tags]] which may refer to any and multiple (or none) of the above
  - [[c]]: ways to categorize resources
  - [[p]]: ways to assert things about resources (mostly abandoned)

In terms of the [[t.km.zettelkasten]] method, this wiki is a commingling of literature notes with something @Sönke-Ahrens warned about: a "personal Wikipedia or a database" [^1]. 

In terms of [[t.tm.getting-things-done]], it's references, someday-maybe, next actions, and public projects.

## Beware!

This is a cluttered cupboard of jumbled jottings riddled with inaccuracies, confusion, and mistakes.

## What This Is Not

- evergreen-notes/digital-garden: for "synthesized" / original-ish publishable big-idea notes that should evolve over time, props to @andy-matuschak and @maggie-appleton
- blog: for "point-in-time" items to be shared: personal news, articles, etc.
  - e.g., [[ar.substack.carpe-noctem]]


[^1]: [[book.how-to-take-smart-notes]]  