TAGOOn+: Generation and Integration of Organizational Ontologies

Abstract

We present TAGOOn+, a tool that supports: 1) the automatic generation of organizational ontologies from models expressed with i*, Tropos and Service-Oriented i*, and 2) the automatic integration of those organizational ontologies with generic or domain ontologies. TAGOOn+1 is a tool that automatically generates organizational ontologies and automates their integration with other ontologies. It has two main purposes: 1) It supports the automatic generation of organizational ontologies from organizational models expressed with i*, Tropos and Service-Oriented i* [2]. To do this, the ontological metamodel for these variants, called OntoiStar+ [3] has been developed. Model-Driven Engineering (MDE) ideas have been applied to transform organizational models into ontologies derived from OntoiStar+. 2) It supports the automatic integration of enriched organizational models with general or domain ontologies [4]. To do this, semantic annotation suggestions [4] are the guidelines to annotate organizational models with concepts coming from ontologies. The iStarML format [1] has been extended with the attribute \sannotation" [5] to store semantic annotations for each model element.

Keywords

• iStar
• organizational modeling
• ontology
• ontology integration
• Model-Driven Engineering

Introduction

which represents the knowledge described in M1; (ii) an ontology Ojoin, which integrates OiStar and OD and represents the knowledge described in M1 and OD; and (iii) a text le TDoc, that describe the OiStar and OD integration process. These ontologies are described in the standard Web Ontology Language OWL. TAGOOn+ has been developed using the environment of the Eclipse project and the Java programming language. It runs on Windows, Linux and Mac. The tool is based on three main modules: Automatic Transformation module. This module implements a transformation process following MDE ideas. Therefore, the ontological metamodel OntoiStar+ has been developed in order to integrate (into an ontology) the i*, Tropos and Service-oriented i* construct de nition [ 3 ]. Moreover, transformation rules have been de ned in [ 2 ] to transform a model M1 into an OiStar ontology. OiStar corresponds to the ontology OntoiStar+ instantiated with individuals that represent the knowledge depicted in M1, including semantic annotations [ 5 ]. We ilustrate the transformation process with a short example. Let's assume that we have an i* based model describing a disease detection process and an ontology OD that describes diseases and the parts of the human body. A task element of the model labeled as Revise esophagus is annotated with the concepts swallow, stomach and animal-organ taken from OD. This task in iStarML format is represented as: <ielement id="01" name="Revise esophagus" type="task" sannotation="swallow stomach animal-organ"/>. After applying the transformation process, the task corresponds to an individual of OiStar (see Fig. 2). Automatic Integration module. This module integrates a semantic annotated model M1 represented as ontology Oistar with the ontology OD used to annotate M1. It parses OD to obtain its hierarchical structure and the description of each concept. Then, each individual of Oistar is related with one or more concepts of OD through links of type is a. In this way, Oistar and OD are integrated in Ojoin. Ojoin contains the knowledge included in the semantically annotated i* based model M1 integrated with the knowledge included in the ontology OD. Following with the example (Fig. 2), the Oistar individual Revise esophagus is related by is a links with the concepts swallow, stomach and animal-organ. Automatic Documentation module. This module generates a text le that describes the Oistar and OD integration process. A fragment of the generated documentation is: type="task" name="Revise esophagus" annotation="swallow" description="To cause a bolus of food or drink to pass from the mouth/throat into the esophagus". The semantic annotations are useful to discover hidden information. In the example, annotations add information to the task about the fact that the esophagus is used to swallow, and it is related with the stomach and it is an organ. Using the Semantic Web Rule Language SWRL, we could generate a simple rule expressing that a person is related with two elements stomach and esophagus, and these element are related too. If this person has problems with its stomach, then is necessary to revise the esophagus. The rule in SWRL would then be: Person(?x) ^ stomach(?y) ^ esophagus(?z) ^ hasRelation(?y,?z) ^ hasProblems(?x,?y) ! hasRevise(?z). This rule is a short example applying reasoning over the organizational knowledge. 3

Conclusion

We have presented the tool TAGOOn+. With this tool, we bring the advantages of ontologies such as querying and reasoning, to the organizational modeling domain. Moreover, as the organizational knowledge is represented in OWL, it could be available to be exploited and consumed in the Semantic Web by paradigms such as Linked Data. On the other hand, we provide the integration of organizational models enriched with semantic annotations with ontologies, which makes organizational knowledge clearer for humans and more accessible to machines. Morever, we believe that a concept which integrates di erent model elements is a strong indicator to implement a new business services inside the organization, improving the understandability and expressiveness of an organizational model.

References

1. C. Cares , X. Franch , A. Perini , and A. Susi . Towards interoperability of i* models using istarml . Computer Standards & Interfaces , 33 ( 1 ): 69 { 79 , 2011 .

2. K. Najera . An ontology-based approach for integrating i* variants . Master's thesis , National Center of Research and Technological Development , Cuernavaca, Morelos, Mexico, 2011 . www.tagoon.semanticbuilder.com/NajeraThesis.pdf.

3. K. Najera , A. Perini , A. Mart nez, and H. Estrada . Supporting i* model integration through an ontology-based approach . In Fifth International i* Workshop (iStar'11) , pages 43 { 48 , 2011 .

4. B. Vazquez . Enriching organizational models through semantic annotation . Master's thesis , National Center of Research and Technological Development , Cuernavaca, Morelos, Mexico, 2012 . www.tagoon.semanticbuilder.com/VazquezThesis.pdf.

5. B. Vazquez , A. Martinez , A. Perini , H. Estrada , and M. Morandini . Enriching Organizational Model through Semantic Annotation . In Proceedings of the Iberoamerican Conference on Electronics Engineering and Computer Science , 2013 .