Towards Semantic Integration of Opinions: Unified Opinion Concepts Ontology and Extraction Task

2025-05-24

AI Summary

Opinion expression exhibits semantic heterogeneity across diverse linguistic formulations, hindering unified modeling. Method: We propose the Unified Opinion Concept (UOC) ontology—the first systematic integration of multi-dimensional opinion semantics (e.g., opinion holder, target, stance, sentiment, intensity)—and formally define the novel task of Unified Opinion Concept Extraction (UOCE), requiring models to output fine-grained opinion triples conforming to UOC structure. We design a semantic-consistency-driven evaluation metric and release UOCE-Bench, a human-annotated benchmark dataset. Our approach combines ontology-guided modeling, semantic parsing, and LLM (LLaMA/Qwen) fine-tuning with prompt engineering. Contribution/Results: We publicly release the first open-source UOC ontology and UOCE-Bench; establish state-of-the-art baselines; and empirically demonstrate that explicit semantic constraints substantially improve both accuracy and interpretability of opinion extraction.

Technology Category

• Natural Language Processing: Information Extraction
• Knowledge Representation and Reasoning: Ontologies
• Data Mining & Knowledge Management: Semantic Web

Application Category

• Web Mining and Content Analysis: Sentiment analysis and opinion mining
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

This paper introduces the Unified Opinion Concepts (UOC) ontology to integrate opinions within their semantic context. The UOC ontology bridges the gap between the semantic representation of opinion across different formulations. It is a unified conceptualisation based on the facets of opinions studied extensively in NLP and semantic structures described through symbolic descriptions. We further propose the Unified Opinion Concept Extraction (UOCE) task of extracting opinions from the text with enhanced expressivity. Additionally, we provide a manually extended and re-annotated evaluation dataset for this task and tailored evaluation metrics to assess the adherence of extracted opinions to UOC semantics. Finally, we establish baseline performance for the UOCE task using state-of-the-art generative models.

Problem

• Integrate opinions within their semantic context using UOC ontology
• Bridge gap between semantic representations of opinions across formulations
• Extract opinions with enhanced expressivity via UOCE task

Innovation

• Unified Opinion Concepts ontology for semantic integration
• Unified Opinion Concept Extraction task with expressivity
• Baseline performance using generative models