Giter Site home page Giter Site logo

monarch-initiative / gpt-mapping-manuscript Goto Github PK

View Code? Open in Web Editor NEW
6.0 4.0 0.0 8.4 MB

Home Page: https://monarch-initiative.github.io/gpt-mapping-manuscript/

License: Creative Commons Attribution 4.0 International

Makefile 0.60% Shell 0.88% HTML 12.09% Jupyter Notebook 86.21% Perl 0.02% Python 0.20%
manubot mapping monarchinitiative ontogpt ontology-alignment

gpt-mapping-manuscript's Issues

Submit mappergpt to OAEI 2023

See sister issue for OAK with the details INCATools/ontology-access-kit#617

Info from email:

The OAEI Bio-ML track will have its 2023 version, for providing a platform for evaluating different ontology alignment systems, especially those using machine learning techniques. We sincerely invite all kinds of OM systems to participate in our Bio-ML track.

In comparison with the 2022 version, we will make the following changes in 2023:

We will set up a new evaluation dataset to test large language models (LLMs) for equivalence matching. The idea is to rank a set of candidate target concepts for a given source concept, by exploring LLMs like Flan-T5 and the GPT-series and prompts.

We will cancel the validation set for simplicity. It will be merged with the testing set in the unsupervised setting, and merged with the training set in the semi-supervised setting.

For subsumption matching, we will cancel the unsupervised setting, but keep the semi-supervised setting, for reducing the workload.

We will use modularization techniques to improve the to-be-matched ontologies that are extracted from the original large-scale ontologies according to the given ground truth mappings. This will keep more of the structure of the original ontology.

As in the last year, we encourage system submission via SEALS or HOBBIT, but can also allow direct result (mappings) submission, which will be marked in our results and reports.

Here are the temporary milestones:
27th July: dataset release
31st August: system submission
30th September: result release
14th October: report release

Organizers: Yuan He, Jiaoyan Chen, Hang Dong, Ernesto Jiménez-Ruiz and Ian Horrocks

@cmungall can we assign this to anyone?

Make sure manuscript deploys on github pages

Right now it does not deploy at all (I did it once manually). Something is this line keeps it from not pushing the updated HTML manuscript to the gh-pages branch.

@hrshdhgd can you take a look of what is happening? If you cant find a solution within 15 minutes, say it here and abandon the task.

Move repo to Monarch?

Or do you want to keep it here? Just so I know which link to include in the paper.

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.