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

TL;DR

This paper presents the UOC ontology for semantically integrating opinions, introduces the UOCE extraction task, and provides a dataset and benchmarks to advance opinion analysis in NLP.

Contribution

The paper introduces the UOC ontology and the UOCE task, offering a unified framework and dataset for extracting opinions with semantic depth.

Findings

Established baseline performance for UOCE using state-of-the-art models.

Provided a manually extended dataset for opinion extraction evaluation.

Proposed evaluation metrics aligned with UOC semantics.

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.