Textual-Based vs. Thinging Machines Conceptual Modeling

2025-06-03

AI Summary

Traditional diagrammatic modeling approaches (e.g., UML) suffer from low efficiency and poor scalability in large-scale conceptual modeling. To address this, this paper proposes a lightweight modeling method grounded in the Thinging Machine (TM) paradigm. It models domain entities as time-aware *thimacs* (thing/machine hybrids), emphasizing two core actions—*create* and *process*—to capture dynamic semantics. The approach integrates multi-level graph abstraction for dimensionality reduction and synergistic text-graph representation to enhance model simplicity and visualization efficacy. Empirical evaluation demonstrates that TM significantly outperforms mainstream diagrammatic methods in semantic clarity, tool operability, and ontology integration ease—particularly in large-scale, highly complex, and dynamically evolving modeling scenarios. This work establishes a novel conceptual modeling paradigm that balances formal rigor, engineering practicality, and scalability.

Technology Category

• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning
• Machine Learning: Graph-based Machine Learning
• Knowledge Representation and Reasoning: Ontologies

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

Software engineers typically interpret the domain description in natural language and translate it into a conceptual model. Three approaches are used in this domain modeling: textual languages, diagrammatic languages, and a mixed based of text and diagrams. According to some researchers, relying on a diagrammatic notation levies certain burdens for designing large models because visual languages are intended to depict everything diagrammatically during a development process but fail to do so for a lack of developer efficiency. It is claimed that textual formats enable easier manipulation in editors and tools and facilitate the integration of ontologies in software systems. In this paper, we explore the problem of the relationship between textual format and diagramming in conceptual modeling. The main focus is modeling based on the so-called thinging machine (TM). Several examples are developed in detail to contrast side-by-side targeted domains represented in textual description and TM modeling. A TM model is defined as a thimac (thing/machine) with a time feature that forms dynamic events over static thimacs utilizing five generic actions: create, process, release, transfer, and receive. This provides a conceptual foundation that can be simplified further by eliminating the actions of release, transfer, and receive. A multilevel reduction in the TM diagram s complexity can also be achieved by assuming diagrammatic notations represent the actions of creation and processing. We envision that special tools will help improve developer efficiency. The study s results of contrasting textual and mix-based descriptions vs. TM modeling justify our claim that TM modeling is a more appropriate methodology than other diagrammatic schemes (e.g., UML classes) examined in this paper.

Problem

• Comparing textual and diagrammatic modeling in conceptual design
• Evaluating efficiency of thinging machine (TM) modeling approach
• Contrasting TM modeling with traditional UML class diagrams

Innovation

• Textual and diagrammatic modeling comparison
• Thinging Machine with five dynamic actions
• Simplified TM diagram complexity reduction