Knowledge Representation for the Development of Collaborative Applications

Abstract

The workflow to develop collaborative applications should be highly adaptable to frequent organizational changes. In order to increase the adaptability of workflow, an ontology-based workflow model is proposed. This model represents the set of steps -along with their order of execution- performed by different entities for developing this kind of applications. Ontology is one of the strategies for the structured representation of a chosen knowledge domain in a formal way, helping to remove ambiguity and redundancy, detecting errors, and allowing automated reasoning. In this work, the ontologies are considered as models to represent knowledge. A case study is presented in order to show the use of Knowledge-based Workflow Model for Collaborative Applications.

-

Ontology provides an ideal solution, for it represents the domain knowledge using description logic symbols, which allows specifying it in a simple, readable way for both humans and machines; as well as performing a much deeper reasoning by means of machines. It facilitates a knowledge base to provide semantic, common understanding, communication, and shared knowledge on the domain of interest, and a knowledge reasoning carrying out an inference process to reach to conclusions on such a base, by means of a reasoner, inference rules, and query languages.

Collaborative applications must provide an appropriate infrastructure to back up group work and support the dynamic structure of the organizations in runtime. In such a way that they can represent inherent knowledge in the applications that support groups of people engaged in a common goal, and provide an interface for a shared environment. This paper, tries to capture the resulting knowledge in the development of such applications so, it proposes an ontology-based workflow model named workflow ontology to develop collaborative applications. This workflow ontology allows representing all the necessary symbols in order to specify the elements for building this type of applications. These symbols make up a knowledge base, which allows to reason and draw conclusions; to generate new knowledge in collaborative domain, using reasoner, inference rules and query languages. The rest of the paper is organized as follows: Section 2, describes briefly the ontology-based knowledge; Section 3, explains the inherent knowledge in the collaborative applications; Section 4, presents the workflow model for collaborative applications and a case study focused on academic virtual space; Section 6, outlines the conclusions and future work. 2

Ontology-based knowledge In recent years, the use of ontologies has extended in diverse areas such as medicine [ 1 ]; bioinformatics [ 2 ]; groupware [ 3, 4 ]; mainly, because they allow a formal explicit specification of a shared conceptualization of certain domain of interest. Conceptualization refers to an abstract model of some knowledge in the world through the identification of relevant concepts of this. Explicit specification, means that the type of concepts used and the constraints on their use are explicitly defined. Formal, reflects the fact that the ontology should be machine-readable. Shared, represents the notion that an ontology captures consensual knowledge that is not reserved to some individual, but it is accepted by a group. So, it is said that ontology establishes the vocabulary used to describe and represent domain knowledge to facilitate machine reasoning. The domain knowledge, describes the main static information and the objects of knowledge [ 5 ]. According to Gruber [ 6 ], domain knowledge in ontologies can be formalized using four kind of components: concepts, relations, axioms and instances.

The OWL representation [ 7 ] facilities are directly based on Description Logics [ 8 ]. This basis confers upon OWL a logical framework, including syntax and modeltheoretical semantics, allowing a knowledge representation language capable of supporting a knowledge base, and a practical, effective reasoning. Protégé is used in order to develop ontologies with OWL to provide graphical interfaces that facilitate the knowledge representation and reasoning. Protégé is an engineering tool open source ontology and a knowledge-based framework, which is widely used due to its scalability and extensibility with lots of plugins; to facilitate inference knowledge through reasoners, query languages, and rules. Therefore, it can be concluded that ontologies are an ideal solution for knowledge representation and reasoning, since it provides a set of symbols through a formal and structured vocabulary. 3

Knowledge in collaborative applications A collaborative application focuses on supporting group work to achieve a common goal by providing a shared workspace. For this reason, the knowledge is the result of a group interaction and its adaptation on application, i.e. depends on:  Group, which carries out a set of tasks to achieve a common goal through a shared environment. Consequently, the group should have a suitable Group Organizational Structure (GOS). This is ruled by the Session Management Policy —SMP— that establishes the same [ 9 ]. The SMP defines a hierarchical or not-hierarchical GOS by means of Roles that users can play. These roles establish the set of Rights/Obligations (R/O) and Status (St) within the group, as well as    the Tasks that can be carried out to accomplish a common goal. The Task is composed of Activities, which use the prevailing shared Resources [ 10 ]. The GOS specification provides the Knowledge about how the interaction among users is carried out; what Roles are involved in the group; what Task is performed by each Role, and what Resources are used to achieve a goal. This Knowledge facilitates the basic elements of a collaborative application.

Interaction, requires: A session, shared workspace where a group will interact. A notification process that provides the users with the necessary information to support the group awareness,—users are aware of another member’s presence in the session, and of the actions that each one of them is carrying out—; and supplies a common context on which the activities of the group are performed, and where the information of the shared resources is stored; thus, creating a group memory to provide understanding and reasoning about the collaborative process. A concurrency process that ensures the consistency of the data being shared; providing collaborating users with dynamically-generated temporary access and permissions, to reduce racing conditions, and to guarantee mutually exclusive resource usage. These permissions depend on the GOS established and on the lock mechanism. So, the interaction knowledge is supplied by concurrency, awareness, and memory group, in such a way that the common context helps avoiding surprises by reducing the probability of conflicts in the group established [ 11 ]. Application, presents the Views that are user interfaces allowing interaction between the users and the application. There are three Views: Information View (IV) which is related to individual information; the Participant View (PV) that is associated with group awareness; and the Context View (CV) which is connected to group memory. The application is made up into phases [ 12 ], where each phase is defined as a global description of the tasks that are active. Therefore, the knowledge is supplied via the IV, PV, and CV, allowing to show the interaction between users and the application; as well as by means of phases which constrain the users who can participate in accordance with the roles they are playing. Adaptation [ 13 ]; adjusts the Views in accordance to changes triggered by notification, in such a manner that the application shows the most recent updates although preserving its functionality. Accordingly, it is necessary to monitor the changes in the session using a detection process; in the case of an adaptable process (when the adaptation is carried out by direct intervention of the user) in a non-hierarchical GOS the pre-adaptation stage is performed. This process accomplishes an agreement, where all users have to reach a consensus on whether an adaptation process should be performed by means of a Voting Tool, which offers several kinds of agreements such as the agreement based on the majority vote, the one based on a maximum or minimum value, etc. In the case of an adaptive process (when the adaptation is automatically performed) or the users have agreed to make the adaptation, an adaptation flow process is performed. When an adaptation flow cannot be completed, a reparation process is invoked, —which returns each component to its previous state— then, users are notified that this adaptation process cannot be achieved. Knowledge is related to modifying one or several application components, in order to change exclusively that part of the application which did not fit the features of the new scenario.

The aforementioned is derived from the architectural model (see Figure 1) presented in [ 10 ] that is the result of the study and review of several tools or frameworks (such as Groupkit [ 14 ], ANTS [ 15 ], and SAGA [ 16 ]), architectures (e.g., Clock [ 17 ], and Clover [ 18 ]), and methodologies (AMENITIES [ 19 ], ClAM [ 3 ], and TOUCHE [ 4 ]). Where the different concepts and terms above mentioned (such as group, role, task, activity, resource, session, notification, concurrency, shared user interface, phase, adaptation, etc.), have been applied to design and develop collaborative applications. All these concepts and terms will be used as symbols, which are part of workflow ontology for the development of collaborative applications.

ADAPTATION Pre-Adaptation Votation

LEVEL Stage Tool

Detection Agreement

Adaptation Stage

Reparation Adaptation Flow

APPLICATION

LEVEL

Phase Information View Participant View Context View

INTERACTION Session

LEVEL

GROUP

LEVEL

Group Organizational Structure

A workflow model for collaborative applications The development of collaborative applications needs to perform a group of coordinated steps which can be accomplished by means of a workflow. The term workflow typically refers to coordinated execution of multiple tasks or operations [ 20 ]. However, workflows lack the expressive power to represent the domain knowledge and the sequence of operations. On the other hand, ontology describes knowledge domain through concepts, relations, axioms and instances, although ontology does not specify how these entities should be used and combined.

Special attention has recently been paid to the development of workflow ontologies, as the workflow can be built on the ontology. The former defines how the tasks or operations should be used and combined, and once that it has been represented, it may be considered a static structure. The latter can symbolize this structure due to its expressive power. So the ontology is an ideal solution for the workflow representation. There are several examples of workflow ontologies: it presents a collaborative workflow for terminology extraction and collaborative modeling of formal ontologies using two tools Protege and OntoLancs [ 21 ]; it allows the development of cooperative and distributed ontologies, based on dependencies management between ontologies modules [ 22 ]; it shows an ontology-based workflows for ontology collaborative development in Protégé [ 23 ],it presents the combination of workflows with ontologies to design way formal protocols for laboratories [ 24 ], it proposes a workflow ontology for the preservation of digital material produced by an organization or a file system [ 25 ]. All these works are focused on building workflow ontologies to represent collaborative work in different areas, however, this paper presents a workflow ontology to develop collaborative application (see Figure 2) using the architectural model elements as the ontology vocabulary.

The Table 1 shows the concepts, relations and axioms of the Workflow Ontology, which establishes; first, the Application is described; second, the SMP name is stated; third, the GOS together with all the elements that contain it are defined; fourth, the Tasks that are part of the Phase and the View are specified; fifth, the Resources that will be presented on the View; sixth, the IV, PV, and CV that will have the View; seventh, the Change produced by Activity; eighth, the Notification and Concurrency (Cc) triggered by Change; ninth, the Resource locked by means of the Cc; tenth, the Adaptation (Ad) generated through Notification; eleventh, the Adaptation is showed on View (IV, PV, and CV); twelfth, it outlines both Views and Phases that are part of the Session, and finally the Sessions that comprise the Application. A case study is an Academic Virtual Space (AVS) which provides the students with a shared workspace to simplify their access to the course material previously loaded by the professor. The AVS includes two roles: 1) The professor can register himself, create groups, upload and download files both his and those of the students, publish and respond to those made by others; and 2) The students, can register themselves in the shared space and access courses, load and unload files (homework and course materials that have been uploaded by the teacher), make publications and to reply to them. Thus, in order to create this workspace only the workflow ontology instances should be defined (see Table 2). The instances are defined in according to the established steps in the workflow ontology. These constitute the knowledge base for developing collaborative applications. Therefore, first, the described Application is AVS; second, the stated SMP is SMP-AVS; third, the defined GOS is GOS-AVS and the elements desdiegxenshaiigtbneistastes isDisplayed triggers

is_showed Figure 2. Workflow Ontology to develop groupware. that contain the GOS-AVS are showed in the Table 2; the fourth, two phases are specified along with their tasks that are part of the Phase, as well as the Views are designated; fifth, the Resources on each View are presented; sixth, the IV, PV, and CV that will have the View are showed; seventh, the produced Change by Activity are revealed in the Notification Colum of the Table 2; eighth, the triggered Cc by Change is indicated by Yes in the Cc column (Y); ninth, the Cc column presents Y, when a Resource is locked; tenth, the generated Adaptation is displayed in the Ad column; eleventh, the showed Adaptation on View is specified with Y in IV, PV, and CV; twelfth, the Views and Phases of the Session are exhibited, and finally the Sessions that comprise the Application. When Table 2 is constructed, and the GOS elements are already defined, it is possible infer, for example: The tasks of each phase and view; the presented resources in each view, the produced changes by each activity, the activated concurrency by change, etc. This inference is very important when a collaborative application has been developed. Furthermore, the AVS knowledge base can be used in another applications, which focus in academic environment and requires two roles (Teacher and Student). In this a manner, it will be possible to reduce time and costs in the development of collaborative applications. 5

Conclusions and Future Work This paper has presented a workflow model for developing collaborative applications using the inherent knowledge of the integrated elements —group, interaction, application and adaptation— necessary to build this applications. These elements are the ontology concepts, and present a set of constrains, that guide and facilitate the development of this applications. Developing a formal and structured knowledge base, which allows building any collaborative application, defining its instances on ontology. Furthermore, in order to simplify and facilitate an ontology description, a specification table is proposed, in such a manner that any individual may be able to develop a collaborative application by means on this table.

The future work will aim to specify a methodology to develop collaborative application, starting with the workflow ontology described in this article.

Tabla 2. Knowledge Base of AVS.

Rol

St

R/O

Task

View (V)

Phase Notification PV CV

References

1. Jovic , A. , Prcela , M, and Gamberger , D. Ontologies in Medical Knowledge Representation . In Proceedings of the 29th Int. Conf. on Information Technology Interfaces , pp. 25 - 28 , ( 2007 )

2. Stevens , R. , Goble , C.A. , and Bechhofer , S. Ontology-based knowledge representation for bioinformatics . Briefings in Bioinformatics , Vol. 1 - 4 , pp. 398 - 414 , ( 2000 )

3. Molina , A.I. , Redondo , M.A. , Ortega , M. , and Hope , U. ClAM. A methodology for the development of groupware user interfaces . Journal of Universal Computer Science ( 2007 )

4. Penichet , V. , Lozano , M. , and Gallud. J. An Ontology to Model Collaborative Organizational Structures in CSCW . In Engineering the User Interface, Springer, pp. 127 - 139 , ( 2008 )

5. Schreiber , G. , Akkermans , H. , Anjewierden , A. , Hoog , R. , Shadbolt , N., Van de Velde , W. , Wielinga , B. Knowledge engineering and management . The KADS Methodology . ( 1999 )

6. Gruber , R. A translation approach to portable ontology specification . Knowledge Acquisition . Vol. 5 , pp. 199 - 220 , ( 1993 )

7. Horrocks , I. , Patel-Schneider , P.F. , and van Harmelen, F. From SHIQ and RDF to OWL: The making of a web ontology language . J. of Web Semantics , Vol. 1 - 1 , pp. 7 - 26 , ( 2003 )

8. Baader , F. , Calvanese , D. , McGuinness , D. , Nardi , D. , and Patel-Schneider , P. F. editors. The Description Logic Handbook: Theory, Implementation and Applications . Cambridge University Press, ( 2003 )

9. Anzures-García , M. , Sánchez-Gálvez , L.A. , Hornos , M.J. , and Paderewski-Rodríguez , P. Ontology-Based Modelling of Session Management Policies for Groupware Applications . Lecture Notes in Computer Science , Vol. 4739 , pp. 57 - 64 , Springer, Heidelberg, ( 2007 )

10. Anzures-García , M. , Sánchez-Gálvez , L.A. , Hornos , M.J. , and Paderewski-Rodríguez , P. SAMCA: Service-based architectural model for collaborative applications . In Proc. 15TH International Conference on Software Engineering and Applications ( 2011 )

11. Anzures-García M. , Sánchez-Gálvez L .A., Hornos M.J. & Paderewski , P. Methontologybased ontology representing a service-based architectural model for collaborative applications . Advances in Soft Computing Algorithms. RCS , Vol. 54 , pp. 77 - 90 , ( 2011 )

12. Anzures-García , M. , Sánchez-Gálvez , L.A. , Hornos , M.J. , and Paderewski-Rodríguez , P. Service-based access control using stages for collaborative systems . Advances in Computer Science and Engineering. Research in Computing Science , Vol. 42 , pp. 311 - 322 , ( 2009 )

13. Anzures-García , M. , Sánchez-Gálvez , L.A. , Hornos , M.J. , and Paderewski , P. Security and adaptability to groupware applications using a set of SOA-based services . Advances in Computer Science and Engineering. RCS , Vol. 45 , pp. 279 - 290 , ( 2010 )

14. Roseman , M. , and Greenberg , S. Building Real-time Groupware with GroupKit, a Groupware ToolKit . ACM Trans. Computer-Human- Interaction , Vol. 3 , 66 - 106 , ( 1996 ).

15. García , P. , and Gómez , A. ANTS Framework for Cooperative Work Environments . IEEE Computer Society Press, Vol. 36 , 3 Los Alamitos, CA, USA, pp. 56 - 62 , ( 2003 ).

16. Fonseca , B. , and Carrapatoso , E. SAGA: A Web Services Architecture for Groupware Applications , In Proc. of the CRIWG, LNCS 4154 , Springer-Verlag, pp. 246 - 261 , ( 2006 ).

17. Graham , T.C.N. , and Urnes , T. Integrating Support for Temporal Media in to an Architecture for Graphical User Interfaces . Proc. of the International Conference on Software Engineering (ICSE'97) , ACM Press, Boston, USA, pp. 172 - 182 , ( 1997 ).

18. Laurillau , Y. , and Nigay , L. Clover Architecture for Groupware . Proc. of the ACM Conference on CSCW. New Orleans , Louisiana, USA, pp. 236 - 245 , ( 2002 )

19. Garrido , J. L. , Gea , M. , Padilla , N. , Canas , J.J. , and Waern , Y. AMENITIES : Modelo de entornos cooperativos . In I. Aedo, P. Díaz & C. Fernández (eds.), Actas del III Congreso Internacional Interacción Persona-Ordenador , pp. 97 - 104 , ( 2002 )

20. Fischer , L. Workflow Handbook . Future Strategeis Inc., Lighthouse Point , FL ( 2004 )

21. Gacitua , R.; Arguello-Casteleiro , M. ; Sawyer , P. ; Des , J. ; Perez, R. ; Fernandez-Prieto , M.J. ; Paniagua, H. , A collaborative workflow for building ontologies: A case study in the biomedical field . Research Challenges in Information Scienc . pp. 121 - 128 ( 2009 )

22. Kozaki , K. , Sunagawa , E. , Kitamura , Y. and Mizoguchi , R. A Framework for Cooperative Ontology Construction Based on Dependency Management of Modules . Vol. 292 of CEUR Workshop Proceedings , pp. 33 - 44 . CEUR-WS.org,( 2007 )

23. Sebastian , A. , Noy , N.F. , Tudorache , T. , Musen , M. A. A Generic Ontology for Collaborative Ontology-Development Workflows . Proceedings of the 16th international conference on Knowledge Engineering: Practice and Patterns , ( 2008 )

24. Maccagnan , A. , Riva , M. , Feltrin , E. , Simionati , B. , Vardanega , T. , Valle , G. , Cannata , N. Combining ontologies and workflows to design formal protocols for biological laboratories , Automated Experimentation , Vol. 2 - 3 , ( 2010 )

25. Mikelakis , M. , Papatheodorou , C. An ontology-based model for preservation workflows . In Proceedings of the 9th International Conference on Digital Preservation , ( 2012 )