OntoAligner-Ensemble: Voting-Based Fusion across Heterogeneous Ontology Alignment Techniques
Year: 2026
Type: software
Abstract: This repository contains the code, baseline runners, and evaluation workflows for OntoAligner-Ensemble, a modular framework for cross-paradigm ontology alignment. We present a controlled evaluation combining predictions from lexical matchers, Knowledge Graph Embedding (KGE) models, and Retrieval-Augmented Generation (RAG) pipelines powered by open-weight and API-based LLMs (including OpenAI, Gemini, and Qwen variants). The framework features a two-stage integration process using voting-based fusion strategies (such as Weighted Voting, Reciprocal Rank Fusion, Condorcet, and Borda Count) paired with post-fusion selection policies to reconcile complementary and conflicting predictions across eight benchmarks from five OAEI tracks. The repository includes baseline and ensemble runners, configuration files, dataset handlers, and output metrics supporting reproducible cross-paradigm ontology alignment experiments.
Language: English
Topic: Semantic Web and Ontologies
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7172111052
