The temporal conceptual data modelling language TREND

TL;DR

The paper introduces TREND, a highly expressive temporal conceptual data modelling language grounded in $\mathcal{DLR}_{US}$, designed to extend ER/UML diagrams with robust temporal constraints. It reports a rigorous, multi-stage evaluation comprising 11 experiments and over 1,000 participants, confirming that a graphical TREND notation is preferred and that extended textual explanations do not markedly improve modelling quality. The findings show that TREND enables novices to understand and produce temporal models with generally solid accuracy, though domain-specific guidance significantly enhances modelling quality. The work highlights the potential of TREND for expressive temporal modelling in practice while outlining directions for tooling, domain-guided training, and ontology-based data access extensions.

Abstract

Temporal conceptual data modelling, as an extension to regular conceptual data modelling languages such as EER and UML class diagrams, has received intermittent attention across the decades. It is receiving renewed interest in the context of, among others, business process modelling that needs robust expressive data models to complement them. None of the proposed temporal conceptual data modelling languages have been tested on understandability and usability by modellers, however, nor is it clear which temporal constraints would be used by modellers or whether the ones included are the relevant temporal constraints. We therefore sought to investigate temporal representations in temporal conceptual data modelling languages, design a, to date, most expressive language, TREND, through small-scale qualitative experiments, and finalise the graphical notation and modelling and understanding in large scale experiments. This involved a series of 11 experiments with over a thousand participants in total, having created 246 temporal conceptual data models. Key outcomes are that choice of label for transition constraints had limited impact, as did extending explanations of the modelling language, but expressing what needs to be modelled in controlled natural language did improve model quality. The experiments also indicate that more training may be needed, in particular guidance for domain experts, to achieve adoption of temporal conceptual data modelling by the community.

Figures (23)

• Figure 1: A: all MADS icons for temporal elements; B: the LTRM clock; C: characteristic graphical adornments of the older models reviewed in Gregersen99.
• Figure 2: A selection of model elements from six different TCDMLs to illustrate the sort of adornments used for temporal constraints in the 'recent' proposals.
• Figure 3: Options for the basic arrow shape for the generic dynamic (transition) constraints and for persistence of those transitions.
• Figure 4: Orchestration of approach to obtain the links between a swappable visual modelling language, the reusable syntax and semantics, and relation to $\mathcal{DLR}_{\mathcal{US}}$, and their relation to the $ER{_{VT}}$ specification in AK08is (equally applicable for $EER{_{VT}^{++}}$). Observe the bi-directional mapping between text and graphics and the unidirectional mapping into $\mathcal{DLR}_{\mathcal{US}}$.
• Figure 5: A small partial Trend diagram (left) and corresponding textual notation (right); see text for explanation.

Theorems & Definitions (4)

• Definition 1: Trend Conceptual Data Model
• Definition 2: Trend Semantics
• Definition 3: Reasoning Services
• Definition 4: Mapping Trend into $\mathcal{DLR}_{\mathcal{US}}$