=Paper=
{{Paper
|id=Vol-1660/ecs-paper2
|storemode=property
|title=Cross-platform Functional Consistency Validation for the Event-Driven Systems: An Ontology-Based Approach
|pdfUrl=https://ceur-ws.org/Vol-1660/ecs-paper2.pdf
|volume=Vol-1660
|authors=Fahim T. Imam
|dblpUrl=https://dblp.org/rec/conf/fois/Imam16
}}
==Cross-platform Functional Consistency Validation for the Event-Driven Systems: An Ontology-Based Approach==
https://ceur-ws.org/Vol-1660/ecs-paper2.pdf
Cross-platform Functional Consistency
Validation for the Event-Driven Systems:
An Ontology-Based Approach
Fahim T. IMAM 1
School of Computing, Queen’s University, ON K7K 3Z5, Canada
fahim.imam@queensu.ca
Introduction. Supporting multiple, heterogeneous platforms for the modern event-
driven systems, e.g., mobile computing based applications, is a common requirement for
any large-scale software engineering projects. When dealing with the functional behav-
ior of these systems, an effective software engineering approach must tackle the follow-
ing key challenges: (a) validating the level of functional consistencies between the cross-
platform application systems; (b) maintaining a consistent co-evolution of the cross-
platform functionalities based on the evolving requirements of the system’s domain; and,
(c) comprehending the functional changes that can be both human comprehensible and
machine processable. The ultimate goal of this work is to develop an approach that can
promote a machine-processable, intelligent decision-support mechanism for the kinds of
functional reasoning that is required in order to mitigate these latter challenges for the
Software Engineers. In order to achieve our goal, we consider the following research
questions to be the critical driving force.
• How to effectively represent the software systems in terms of their functional
behavior that can be both human comprehensible and machine processable?
• Can we incorporate the AI-based commonsense reasoning mechanisms for the
functional behavior reasoning? What are the exact characteristics of the functional
reasoning that is suitable for our required consistency validation mechanism?
Key Motivation. Despite the unavoidable heterogeneity issues with multiple plat-
forms, the systems developed for those platforms must confirm that their intended set
of functionalities are consistently implemented across the platforms. The tasks of main-
tenance in these systems can be quite expensive and challenging. An effective mainte-
nance for the large-scale evolving systems must require a rigorous strategy for consis-
tency management and change propagation across the overall system components. While
the importance of software testing cannot be denied as part of the maintenance process,
generating all possible test cases, and tracing the solution fragment for negative results,
can be overwhelming. Contrary to software testing, we considered a more systematic,
formal approach for the consistency management. However, our approach can help cat-
egorizing the set of test cases that must be executed in order to validate the functional
consistencies between the cross-platform and evolving systems.
1 The author of this research is supervised by Dr. Thomas R. Dean from the Department of Electrical and
Computer Engineering at Queen’s University, Canada. This research is supported by the NSERC, Canada.
Methodology. The idea is to utilize the powerful notion of ontology in order to mit-
igate the heterogeneity issues among the systems that are intended to have a set of identi-
cal functional goals. By representing the functional behavior of the systems into an ontol-
ogy, we can have an effective mechanism to validate the levels of functional consistency
among the systems. Since we are dealing with the event-based systems like mobile apps,
the focus of the representation should be the functionalities that can be observed from
its User Interface (UI) elements. Since the events associated with the UI elements are
indicative of the system’s behavior [1], these elements should provide the key source of
knowledge regarding the functional model of the system. A set of such functional mod-
els should be captured at a level of abstraction that can be useful for our purpose. Once
we have an ontology that can represent the functional models of the systems, different
source code components of the involving systems can be annotated and mapped with
that ontology. The ontological models will then represent a set of language-independent,
comparable functionality models for each of the source systems. Next, the functionality
models should be mapped and merged through the ontology. The ontology reasoner can
then derive the implicit classification of functionalities, i.e., the logical functional de-
composition of the source systems. Finally, as part of the reasoning process, the ontology
reasoner would be able to detect the inconsistent set of functional features, along with
the logical reasoning. To our best knowledge, there exist no direct, adequate research
contributions that can be considered comparable to our approach.
Results. Based on the theory of action, events, and change, as understood from
the literature of commonsense reasoning [2], event calculus, and functional reasoning
[3], we have developed an ontology called the Event-based Functional Behavior Ontol-
ogy (EFBO, http://cs.queensu.ca/~imam/uifo.html) for our purpose. The core
classes of the EFBO are the Event, Agent, and the Interface classes. All the other entities
within the EFBO are logically based on these three classes. The concept of Event within
the EFBO essentially corresponds to the theory of SPAN [4], as endorsed by the Basic
Formal Ontology (BFO). The concept of Agent and Interface, on the other hand, corre-
spond to the theory of SNAP [4] within the BFO. The EFBO provides an effective repre-
sentational facility to model the functional behavior of any event-based systems that can
be comprehensible for both humans and machines. We have some preliminary results on
modelling different systems using the EFBO. We are currently in the stage of annotating
and mapping a number of cross-platform mobile apps that can be used to validate their
functional consistencies based on the EFBO. We are also in the process of implementing
a prototype application that can utilize our cross-platform functional validation approach
in a semi-automated manner. The next step would be to develop a comparison metrics to
verify the effectiveness of our approach.
References
[1] D. M. Hilbert and D. F. Redmiles. Extracting usability information from user interface events. ACM
Computing Surveys (CSUR), 32(4):384–421, 2000.
[2] E. Davis and G. Marcus. Commonsense reasoning and commonsense knowledge in artificial
intelligence. Communications of the ACM, 58(9):92–103, 2015.
[3] B. H. Far and A. H. Elamy. Functional reasoning theories: Problems and perspectives. AIE EDAM,
19(02):75–88, 2005.
[4] P. Grenon and B. Smith. SNAP and SPAN: Towards dynamic spatial ontology. Spatial cognition and
computation, 4(1):69–104, 2004.