README.md

metadata

license: mit
language:
  - en
tags:
  - OntoLearner
  - ontology-learning
  - chemistry
pretty_name: Chemistry

Chemistry Domain Ontologies

Overview

The chemistry domain encompasses the structured representation and formalization of chemical knowledge, including entities, reactions, processes, and methodologies. It plays a critical role in knowledge representation by enabling the integration, sharing, and computational analysis of chemical data across diverse subfields such as organic, inorganic, physical, and computational chemistry. This domain facilitates the advancement of scientific research and innovation by providing a standardized framework for the precise and interoperable exchange of chemical information.

Ontologies

Ontology ID	Full Name	Classes	Properties	Last Updated
AFO	Allotrope Foundation Ontology (AFO)	3871	318	2024-06-28
ChEBI	Chemical Entities of Biological Interest (ChEBI)	220816	10	01/01/2025
CHEMINF	Chemical Information Ontology (CHEMINF)	358	52	None
CHIRO	CHEBI Integrated Role Ontology (CHIRO)	13930	15	2015-11-23
ChMO	Chemical Methods Ontology (ChMO)	3202	27	2022-04-19
FIX	FIX Ontology (FIX)	1163	5	2020-04-13
MassSpectrometry	Mass Spectrometry Ontology (MassSpectrometry)	3636	12	12:02:2025
MOP	Molecular Process Ontology (MOP)	3717	11	2022-05-11
NMRCV	Nuclear Magnetic Resonance Controlled Vocabulary (NMRCV)	757	0	2017-10-19
OntoKin	Chemical Kinetics Ontology (OntoKin)	83	136	08 February 2022
PROCO	PROcess Chemistry Ontology (PROCO)	970	61	04-14-2022
PSIMOD	Proteomics Standards Initiative (PSI) Protein Modifications Ontology (PSI-MOD)	2098	4	2022-06-13
REX	Physico-chemical process ontology (REX)	552	6	2025-03-11
RXNO	Reaction Ontology (RXNO)	1109	14	2021-12-16
VIBSO	Vibrational Spectroscopy Ontology (VIBSO)	598	53	2024-09-23

Dataset Files

Each ontology directory contains the following files:

1. <ontology_id>.<format> - The original ontology file
2. term_typings.json - A Dataset of term-to-type mappings
3. taxonomies.json - Dataset of taxonomic relations
4. non_taxonomic_relations.json - Dataset of non-taxonomic relations
5. <ontology_id>.rst - Documentation describing the ontology

Usage

These datasets are intended for ontology learning research and applications. Here's how to use them with OntoLearner:

First of all, install the OntoLearner library via PiP:

pip install ontolearner

How to load an ontology or LLM4OL Paradigm tasks datasets?

from ontolearner import AFO

ontology = AFO()

# Load an ontology.
ontology.load()  

# Load (or extract) LLMs4OL Paradigm tasks datasets
data = ontology.extract()

How use the loaded dataset for LLM4OL Paradigm task settings?

# Import core modules from the OntoLearner library
from ontolearner import AFO, LearnerPipeline, train_test_split

# Load the AFO ontology, which contains concepts related to wines, their properties, and categories
ontology = AFO()
ontology.load()  # Load entities, types, and structured term annotations from the ontology
data = ontology.extract()

# Split into train and test sets
train_data, test_data = train_test_split(data, test_size=0.2, random_state=42)

# Initialize a multi-component learning pipeline (retriever + LLM)
# This configuration enables a Retrieval-Augmented Generation (RAG) setup
pipeline = LearnerPipeline(
    retriever_id='sentence-transformers/all-MiniLM-L6-v2',      # Dense retriever model for nearest neighbor search
    llm_id='Qwen/Qwen2.5-0.5B-Instruct',                        # Lightweight instruction-tuned LLM for reasoning
    hf_token='...',                                             # Hugging Face token for accessing gated models
    batch_size=32,                                              # Batch size for training/prediction if supported
    top_k=5                                                     # Number of top retrievals to include in RAG prompting
)

# Run the pipeline: training, prediction, and evaluation in one call
outputs = pipeline(
    train_data=train_data,
    test_data=test_data,
    evaluate=True,              # Compute metrics like precision, recall, and F1
    task='term-typing'          # Specifies the task
                                # Other options: "taxonomy-discovery" or "non-taxonomy-discovery"
)

# Print final evaluation metrics
print("Metrics:", outputs['metrics'])

# Print the total time taken for the full pipeline execution
print("Elapsed time:", outputs['elapsed_time'])

# Print all outputs (including predictions)
print(outputs)

For more detailed documentation, see the

Citation

If you find our work helpful, feel free to give us a cite.

@inproceedings{babaei2023llms4ol,
  title={LLMs4OL: Large language models for ontology learning},
  author={Babaei Giglou, Hamed and D’Souza, Jennifer and Auer, S{\"o}ren},
  booktitle={International Semantic Web Conference},
  pages={408--427},
  year={2023},
  organization={Springer}
}