[Paper Review] An Ontology of Preference-Based Multiobjective Metaheuristics

Longmei Li, Iryna Yevseyeva+4 more|arXiv (Cornell University)|Sep 26, 2016

Advanced Multi-Objective Optimization Algorithms118 references3 citations

TL;DR

This paper proposes the first OWL-based ontology for preference-based multi-objective metaheuristics (PMOMH), systematically modeling algorithms, preference models, and their integration. Built using Protégé, the ontology enables structured knowledge management, query support, and discovery of research gaps through formal, machine-readable representation of PMOMH methods and their relationships.

ABSTRACT

User preference integration is of great importance in multi-objective optimization, in particular in many objective optimization. Preferences have long been considered in traditional multicriteria decision making (MCDM) which is based on mathematical programming. Recently, it is integrated in multi-objective metaheuristics (MOMH), resulting in focus on preferred parts of the Pareto front instead of the whole Pareto front. The number of publications on preference-based multi-objective metaheuristics has increased rapidly over the past decades. There already exist various preference handling methods and MOMH methods, which have been combined in diverse ways. This article proposes to use the Web Ontology Language (OWL) to model and systematize the results developed in this field. A review of the existing work is provided, based on which an ontology is built and instantiated with state-of-the-art results. The OWL ontology is made public and open to future extension. Moreover, the usage of the ontology is exemplified for different use-cases, including querying for methods that match an engineering application, bibliometric analysis, checking existence of combinations of preference models and MOMH techniques, and discovering opportunities for new research and open research questions.

Motivation & Objective

• To address the growing complexity and fragmentation in preference-based multi-objective metaheuristics (PMOMH) research.
• To provide a formal, standardized, and extensible knowledge representation for PMOMH methods and their integration with multicriteria decision making (MCDM).
• To support researchers in discovering suitable methods for applications, identifying research gaps, and enabling bibliometric analysis.
• To create a shared, extensible, and machine-processable framework for future collaboration across MCDM and MOMH communities.
• To facilitate the integration of new algorithms, preference models, and practical applications into a unified knowledge system.

Proposed method

• The authors developed a comprehensive OWL ontology using the Protégé framework to model core concepts in PMOMH, including MOMH types, preference models, and integration techniques.
• The ontology encodes 13 dominance relations, 12 preference models, and 15 MOMH algorithms, with formal axioms and hierarchical relationships.
• It supports instantiation with state-of-the-art PMOMH methods, enabling formal reasoning and querying over the knowledge base.
• The ontology is hosted in WebProtégé for collaborative editing, version control, and extensibility by the research community.
• Use cases such as method matching, bibliometric analysis, and research gap detection are demonstrated through concrete queries and examples.
• The ontology is designed as an evolving system, open to contributions from MCDM, MOMH, and application domains.

Experimental results

Research questions

• RQ1 How can the growing diversity of PMOMH methods and preference integration techniques be systematically organized and represented?
• RQ2 What are the key relationships between preference models, dominance relations, and MOMH algorithms in PMOMH?
• RQ3 How can formal ontologies improve knowledge discovery, method selection, and research planning in PMOMH?
• RQ4 Which combinations of preference models and MOMH techniques are currently underexplored or missing in the literature?
• RQ5 How can the ontology support the identification of new research opportunities and open problems in PMOMH?

Key findings

• The proposed PMOMH ontology is the first formal, OWL-based knowledge model for preference-based multi-objective metaheuristics, enabling structured and machine-processable representation of the field.
• The ontology successfully models 15 MOMH algorithms, 12 preference models, and 13 dominance relations, with formal axioms and hierarchical relationships.
• Use cases demonstrate the ontology's utility in querying for method-application matches, performing bibliometric analysis, and identifying missing method combinations.
• The ontology enables discovery of research gaps, such as the lack of integration between reference point-based methods and interactive MCDM techniques like the reference point method.
• The system supports dynamic extension, with WebProtégé enabling collaborative updates, comments, and contributions from researchers across MCDM and MOMH communities.
• The ontology is publicly available and designed to grow with future research, enhancing knowledge sharing and reducing redundancy in method development.