With the growing focus on business analytics and data-driven decision-making, there is a greater need for humans to interact effectively with data. We propose that presenting data to human users in terms of instances and attributes provides a more flexible and usable structure for querying, exploring, and analysing data. Compared to a traditional representation, an instance-based representation does not impose any predefined classification schema over the data when it is presented to users. This paper examines the potential utility of instance-based data through two laboratory experiments – the first focusing on exploration of data for pattern discovery (open-ended tasks) and the second on retrieval of information (closed-ended tasks). In both cases, participants were able to achieve better results in tasks using instance-based data than using class-based representations. Given the growing need for self-service analytics, as well as using information for purposes not anticipated when it was collected, we show that instance-based representations can be an effective way to satisfy the emerging needs of information users.

Information systems are intended to be faithful accounts of real-world applications. As an integral part of the development process, analysts create conceptual models in order to understand the application and communicate requirements. Failure to do so has been a prominent reason for IT projects' failure. Hence, improving the quality of models could have a major impact on the information systems' success. To guide the modeling process, researchers use ontology to create more expressive representations of reality. However, improving expressiveness can make the models complicated and cause cognitive hurdles for users. Therefore, the question is whether ontological guidance is worth the trade-off between expressiveness and complexity. This paper describes a meta-analysis of empirical research examining the impact of ontological guidance on users' understandability. The results show that ontological guidance can improve users' understanding of conceptual models, especially those requiring deeper understanding, thus providing support for ontological guidance in conceptual modeling.

The idea that ontological theories could be used to provide a theoretical basis for evaluating conceptual-modeling grammars and scripts was initiated some thirty years ago. The authors provide some brief history about the early research that discussed how ontological theories might be used to study conceptual-modeling grammars and scripts. They then give an overview of the progress that has been made using various elements of this early research. Next, they outline some of the major critiques that have been made of conceptual-modeling work that uses ontological foundations. Finally, the authors suggest some areas that they believe hold substantial promise in using ontology to better understand and predict conceptual-modeling phenomena.

Adoption of Software Product Line Engineering (SPLE) to support systematic reuse of software-related artifacts within product families is challenging, time-consuming and error-prone. Analyzing the variability of existing artifacts needs to reflect different perspectives and preferences of stakeholders in order to facilitate decisions in SPLE adoption. Considering that requirements drive many development methods and activities, we introduce an approach to analyze variability of behaviors as presented in functional requirements. The approach, called semantic and ontological variability analysis (SOVA), uses ontological and semantic considerations to automatically analyze differences between initial states (preconditions), external events (triggers) that act on the system, and final states (post-conditions) of behaviors. The approach generates feature diagrams typically used in SPLE to model variability. Those diagrams are organized according to perspective profiles, reflecting the needs and preferences of the potential stakeholders for given tasks. We conducted an empirical study to examine the usefulness of the approach by comparing it to an existing tool which is mainly based on a latent semantic analysis measurement. SOVA appears to create outputs that are more comprehensible in significantly shorter times. These results demonstrate SOVA's potential to allow for flexible, behavior-oriented variability analysis.

Conceptual models are used to support understanding of and communication about application domains in information systems development. Such models are created using modeling grammars (usually employing graphic representation). To be effective, a grammar should support precise representation of domain concepts and their relationships. Ontology languages such as OWL emerged to define terminologies to support information sharing on the Web. These languages have features that enable representation of semantic relationships among domain concepts and of domain rules, not readily possible with extant conceptual modeling techniques. However, the emphasis in ontology languages has been on formalization and being computer-readable, not on how they can be used to convey domain semantics. Hence, it is unclear how they can be used as conceptual modeling grammars. We suggest using philosophically based ontological principles to guide the use of OWL as a conceptual modeling grammar. The paper presents specific guidelines for creating conceptual models in OWL and demonstrates, via example, the application of the guidelines to creating representations of domain phenomena. To test the effectiveness of the guidelines we conducted an empirical study comparing how well diagrams created with the guidelines support domain understanding in comparison to diagrams created without the guidelines. The results indicate that diagrams created with the guidelines led to better domain understanding of participants.

A common way to represent organizational domains is the use of business process models. A Workflow-net (WF-net) is an application of Petri Nets (with additional rules) that model business process behavior. However, the use of WF-nets to model business processes has some shortcomings. In particular, no rules exist beyond the general constraints of WF-nets to guide the mapping of an actual process into a net. Syntactically correct WF-nets may provide meaningful models of how organizations conduct their business processes. Moreover, the processes represented by these nets may not be feasible to execute or reach their business goals when executed. In this paper, the authors propose a set of rules for mapping the domain in which a process operates into a WF-net, which they derived by attaching ontological semantics to WF-nets. The rules guide the construction of WF-nets, which are meaningful in that their nodes and transitions are directly related to the modeled (business) domains. Furthermore, the proposed semantics imposes on the process models constraints that guide the development of valid process models, namely, models that assure that the process can accomplish its goal when executed.

An important step in developing the requirements for an information system is analyzing the application domain. In this step, conceptual models are used for representing an application domain. However, while languages for software design are available and widely used, no generally accepted language exists for conceptual modeling. This work suggests the use of object-oriented software modeling languages also for conceptual modeling. Such use can support a more accurate transition from domain models to software models. As software-modeling languages were not intended for modeling application domains, their constructs lack the required semantics. While previous papers addressed the representation of structural elements of domains using object concepts, this paper addresses behavioral aspects, related to change and interaction. The proposed semantics are based on a mapping between ontological concepts that describe behavior and object-oriented constructs related to dynamics. Based on these mappings, modeling rules are proposed to guide the modeler in creating ontologically well-formed models. The mappings and rules are exemplified using UML and are demonstrated using a case study.

The Unified Modeling Language (UML) is widely accepted as the de-facto standard for object-oriented information systems (IS) design and software modeling. Recent research has proposed to extend the use of UML to conceptual modeling of application domains. Conceptual models serve both as the basis for communication and domain understanding among analysts, and as the starting point for IS software design. Prior research has proposed a set of modeling rules to provide the analyst or modeler with guidelines for using UML for conceptual modeling.This paper provides an empirical assessment of the benefits derived from using these modeling rules. Using an experimental study involving 53 subjects, it examines the effects of different levels of modeling rule application on application domain understanding. The results show statistically significant differences in the level of domain understanding that was derived from the various models.

Information systems are situated in and are representations of some business or organizational domain. Hence, understanding the application domain is critical to the success of information systems development. To support domain understanding, the application domain is represented in conceptual models. The correctness of conceptual models can affect the development outcome and prevent costly rework during later development stages. This paper proposes a method to restrict the syntax of a modeling language to ensure that only possible configurations of a domain can be modeled, thus increasing the likelihood of creating correct domain models. The proposed method, based on domain ontologies, captures relationships among domain elements via constraints on the language metamodel, thus restricting the set of statements about the domain that can be generated with the language. In effect, this method creates domain specific modeling languages from more generic ones. The method is demonstrated using the Unified Modeling Language (UML). Specifically, it is applied to the subset of UML dealing with object behavior and its applicability is demonstrated on a specific modeling example.

Database design commonly assumes, explicitly or implicitly, that instances must belong to classes. This can be termed the assumption of inherent classification . We argue that the extent and complexity of problems in schema integration, schema evolution, and interoperability are, to a large degree, consequences of inherent classification. Furthermore, we make the case that the assumption of inherent classification violates philosophical and cognitive guidelines on classification and is, therefore, inappropriate in view of the role of data modeling in representing knowledge about application domains. As an alternative, we propose a layered approach to modeling in which information about instances is separated from any particular classification. Two data modeling layers are proposed: (1) an instance model consisting of an instance base (i.e., information about instances and properties) and operations to populate, use, and maintain it; and (2) a class model consisting of a class base (i.e., information about classes defined in terms of properties) and operations to populate, use, and maintain it. The two-layered model provides class independence . This is analogous to the arguments of data independence offered by the relational model in comparison to hierarchical and network models. We show that a two-layered approach yields several advantages. In particular, schema integration is shown to be partially an artifact of inherent classification that can be greatly simplified in designing a database based on a layered model; schema evolution is supported without the complexity of operations currently required by class-based models; and the difficulties associated with interoperability among heterogeneous databases are reduced because there is no need to agree on the semantics of classes among independent databases. We conclude by considering the adequacy of a two-layered approach, outlining possible implementation strategies, and drawing attention to some practical considerations.

Conceptual models or semantic data models were developed to capture the meaning of an application domain as perceived by someone. Moreover, concepts employed in semantic data models have recently been adopted in object-oriented approaches to systems analysis and design. To employ conceptual modeling constructs effectively, their meanings have to be defined rigorously. Often, however, rigorous definitions of these constructs are missing. This situation occurs especially in the case of the relationship construct. Empirical evidence shows that use of relationships is often problematical as a way of communicating the meaning of an application domain. For example, users of conceptual modeling methodologies are frequently confused about whether to show an association between things via a relationship, an entity, or an attribute. Because conceptual models are intended to capture knowledge about a real-world domain, we take the view that the meaning of modeling constructs should be sought in models of reality. Accordingly, we use ontology, which is the branch of philosophy dealing with models of reality, to analyze the meaning of common conceptual modeling constructs. Our analysis provides a precise definition of several conceptual modeling constructs. Based on our analysis, we derive rules for the use of relationships in entity-relationship conceptual modeling. Moreover, we show how the rules resolve ambiguities that exist in current practice and how they can enrich the capacity of an entity-relationship conceptual model to capture knowledge about an application domain.

A method for automating the process of system decomposition is described. The method is based on a formal specification scheme, formal definition of good decomposition, heuristic rules governing the search for good candidate decompositions, and a measure of complexity that allows ranking of the candidate decompositions. The decomposition method has been implemented as a set of experimental computerized systems analysis tools and applied to a standard problem for which other designs already exist. The results are encouraging, in that decompositions generated using other methodologies map easily into those suggested by the computerized tools. Additionally, the use of the method indicates that when more than one 'good' decomposition is suggested by the system, the specifications might have been incomplete. That is, the computerized tools can identify areas where more information should be sought by analysis.>

An ontological model of an information system that provides precise definitions of fundamental concepts like system, subsystem, and coupling is proposed. This model is used to analyze some static and dynamic properties of an information system and to examine the question of what constitutes a good decomposition of an information system. Some of the major types of information system formalisms that bear on the authors' goals and their respective strengths and weaknesses relative to the model are briefly reviewed. Also articulated are some of the fundamental notions that underlie the model. Those basic notions are then used to examine the nature and some dynamics of system decomposition. The model's predictive power is discussed.>