Kiran Prakash Sawant

An efficient XML to OWL converter

XML has become the de-facto standard of data exchange format in E-businesses. Although XML can support syntactic inter-operability, problems arise when data sources represented as XML documents are needed to be integrated. The reason is that XML lacks support for efficient sharing of conceptualization. The Web Ontology Language (OWL) can play an important role here as it can enable semantic inter-operability, and it supports the representation of domain knowledge using classes, properties and instances for applications. In many applications it is required to convert huge XML documents automatically to OWL ontologies, which is receiving a lot of attention. There are some existing converters for this job. Unfortunately they have serious shortcomings, e. g., they do not address the handling of characteristics like internal references, (transitive) import(s), include etc. which are commonly used in XML Schemas. To alleviate these drawbacks, we propose a new framework for mapping XML to OWL automatically. We illustrate our technique on examples to show the efficacy of our approach. We also provide the performance measures of our approach on some standard datasets. We also check the correctness of the conversion process.

Ontology modeling of UBL process diagrams using OWL

We present a logical framework for modeling Universal Business Language (UBL) processes. The proposed framework provides a representation of the dynamic world being modeled on the user supplied axioms about preconditions and the initial state of the world, and produces a workflow specification at a higher level of abstraction. We use the Frieghtbilling process of the UBL as a case study for our experiment. Further we extract ontology from the associated UBL document that can ensure efficient information retrieval, discovery and auditing. We use the popular semantic web formalism, Web Ontology Language (OWL) for ontology construction purposes. These domain ontologies can play a useful role in many applications, like constraining requirements.

Deriving requirements model from textual use cases

In this paper, we present an approach to derive structured requirements models from textual use case requirements in the form of process diagrams and ontology using methods based on computational linguistics. These proposed requirements models are capable of modeling both structural and behavioral entities present in a use case. We consider a corpus containing 123 actual requirements use cases created by Infosys Ltd. and translated them to process diagrams and ontology. For evaluating the performance of conversion we propose a few metrics and show that on average our linguistic engine mis-identied 2% of actions and missed out only 3% of the actions described in the input text.

Enforcing structure on textual use cases via annotation models

Textual use cases are commonly used to represent software requirements at initial stages. However in most of the cases, these documents are unstructured. In this paper, we present a linguistic engine for processing textual use cases and extract a structured model in terms of an annotation model out of these use cases. An annotation model of a use case can further be used to generate various UML requirements models, Business Process Models and ontology. The implementation details of Natural Language Processing (NLP) technique employed by us for the linguistic engine is described in this paper in detail. Also, we consider a corpus containing 123 use cases from real-life industrial projects within our company, and translate them into annotation models using our NLP technique. For evaluating the performance of conversion we use a few metrics and report some promising results for our linguistic engine.

Generating topic maps from XML/XSD documents

Over the years, XML has become a prominent standard for data exchange on the Web. Although XML can support syntactic inter-operability, problems arise when data sources represented as XML documents are needed to be integrated. The reason is that XML does not support sharing of concepts. This issues can be overcome by having suitable ontology representations of XML documents. The Semantic Web languages can play an important role here as they support ontology representation. Topic Map is a Semantic Web formalism (an ISO standard) for knowledge representation and interchange with an emphasis on navigation and retrieval of information. In this work we propose a framework for automatically generating Topic Map ontology from XML. We make use of XQuery and Tolog to validate our method. We illustrate our technique on examples. We also provide performance measures of our approach on some standard data sets.

A knowledge-based formalization of UBL processes using hybrid programs

In business process management, it has become necessary to semantically annotate business processes with ontology for better service provisioning, integration and composition. The OASIS has introduced Universal Business Languages (UBL) as a common format for standardizing common business documents as well as business processes. We propose a method for capturing and managing the information stored in business transactions based on the widely used UBL. We use hybrid programs as the knowledge representation formalism. The formalism of hybrid programs integrates normal clauses (using the parlance of logic programming) with ontology specified in OWL-DL (semantic web standard). This hybrid system uses XSB Prolog for rule reasoning and Pellet for ontology reasoning. We have built a tool that takes a description of a UBL process as input and produces the corresponding hybrid program as output knowledge base on which suitable queries can be fired.