Title: Convergence of Knowledge Graph, Ontologies, and Machine Learning for Reasoning and Decision Making


In the last decade, machine learning algorithms have revolutionized a myriad of human activities, showcasing remarkable progress and becoming integral to technological advancement. As they evolve, these algorithms are increasingly embedded in various domains, heralding a new era of innovation and efficiency. However, their journey towards becoming ubiquitous tools in our digital repertoire faces significant obstacles. Issues such as the opacity of algorithmic decisions, challenges in model integration and knowledge transfer, suboptimal performance in sparse data environments, and struggles with complex data structures, pose critical hurdles. These challenges undermine the trust in and effectiveness of machine learning models, limiting their practical application and the realization of their full potential.

Scope and Purpose

The primary aim of this working group is to address these pivotal challenges by embarking on a comprehensive exploration of the synergy between knowledge graphs, ontologies, and machine learning models. By focusing on Technology Readiness Levels (TRL) 1 to 6, we intend to foster a deeper integration of these elements, significantly elevating both the trustworthiness and the operational performance of machine learning systems. Our research will scrutinize key models such as Generative AI, GraphML, and Hybrid AI, with a particular emphasis on developing hybrid KG-O-ML architectures that seamlessly blend knowledge graphs and ontologies with machine learning to bolster decision-making processes.

In collaboration with the Industrial Ontology Foundry and its Production Planning and Scheduling Working Group, this initiative aspires to craft a cohesive framework that not only advances the state of the art in machine learning but also democratizes the benefits of these technologies across diverse sectors. The knowledge and insights garnered will be disseminated through various channels to enrich the broader community and stimulate further innovation.

Our exploration will span a wide array of applications, from Digital Twin technologies and the Internet of Things (IoT) to Data Harmonization and Model-Based Engineering, with a keen interest in fields like additive manufacturing, biomanufacturing, healthcare, and recommender systems. Through this multidisciplinary effort, we aim to carve a path toward overcoming the existing barriers, unlocking new possibilities for machine learning applications, and paving the way for their more effective and trusted use in society.

Expected outcome of the WG

A review paper on surveying the state-of-the-art development and research gaps in integrating knowledge graphs, ontologies, and machine learning models; a special session in the internationally renowned conference, such as ICPR, ASME’s IDETC/CIE, etc., and a special issue in journals, such as ASME’s JCISE, IJCIM, JIM; this WG also plan to develop different channels, including knowledge sharing platform, to share the findings and knowledge among the affiliated members.

Tentative schedule

Bi-weekly meetings; a draft of the literature review paper in about 6 months; a call-for-paper in 3-6 months. It will be a continuous group, but the proposed outcomes are planned for one year.

Member relationship

Members are expected to provide input to surveys, peer review of ongoing development, and collaborate in research. 

Contributors will receive credit as co-authors in published review papers and know what is happening in the area. Members are also encouraged to become active participants as co-organizer or co-guest editors for the symposiums and special issues.

Introductory Video

Join the discussion at:


Zhaohui Geng

Assistant Professor in Industrial and Systems Engineering, Ohio University

Bernadin Namoano

Lecturer in Digital twin and machine learning Centre, Cranfield University

Dusan Sormaz

Professor Emeritus at Ohio University