Irena Holubová

On Change Detection of XML Schemas

Change detection of XML data has emerged as an important research issue in the last decade, however the majority of change detection algorithms focuses on XML documents rather than schemas. This is because documents that contain data are deemed more significant than the schema itself. This paper looks at the problem from a different perspective by maintaining XML schema (XSD) changes and providing a more meaningful description of the detected changes. Our proposed algorithm XS-Diff uses the technique of storing XML Schema versions in a relational database where the detection and storage of delta changes are employed on relational tables. We demonstrate the correctness of the proposed algorithm through a set of synthetic data. Also, our experimental results show that XS-Diff, is a more meaningful method than other change detection methods for providing deltas that are optimal or near-optimal and semantically correct.

Linked Open Data for Legislative Domain - Ontology and Experimental Data

In this paper, we present an approach of publishing legislative content as Linked Open Data (LOD). LOD is a set of principles of publishing data on the Web in a machine-readable way so that links between different data sets, possibly published by different publishers, can be created. Therefore, LOD enable not only to publish data but also enrich with other existing data published according to the principles. We present what is the motivation for publishing legislation as LOD and what benefits can be gained. We then introduce a legislative ontology which builds on existing commonly used ontologies. We also show how we converted existing sources of legislation in Czech Republic to LOD.

REST API Management and Evolution Using MDA

Nowadays, the REST (Representational State Transfer) is becoming the most commonly used way how to create, publish, and consume Web Services, exploiting JSON (JavaScript Object Notation) as the data exchange format. However, there are still no official standards for REST (like WSDL for Web Services) and for JSON (like XML Schema for the XML language). Next, there are no defined instructions how the resource versions should be managed. In this paper we focus on managing and generating REST resources based on the MDA (Model-Driven Architecture) principle which enables designing and maintenance of complex projects. We introduce a way how to describe the REST API in MDA and how to provide automatic evolution management between subsequent API versions derived from the original version. The proposed solution describes a novel model which represents REST requests and algorithms for providing evolution of this model based on changes done in the PIM (Platform-Independent Model) of MDA.

XS-Diff: XML schema change detection algorithm

Detecting changes in XML data has emerged as an important research issue in the last decade, but the majority of change detection algorithms focus on XML documents rather than on their schemas because documents that contain data are deemed more significant than the schema itself. However, the XML schema change detection tool is essential, especially in situations where we need to maintain related XML documents with evolving schema, sustain relational schema generated by schema-conscious approach for storing XML data and provide support for XML versioning. This paper focuses on XML Schema XSD changes and provides a more meaningful description of the detected changes. Our proposed algorithm XS-Diff uses the technique of storing XML Schema versions in a relational database where the detection and storage of delta changes are employed on relational tables. We demonstrate the correctness of the proposed algorithm through both synthetic and real data sets without deteriorating the execution time.

DaemonX: Design, Adaptation, Evolution, and Management of Native XML (and More Other) Formats

The most common applications of the todays IT world are information systems. The problems related to their design and implementation have sufficiently been solved. However, the true problems occur when an IS is already deployed and user requirements change. Currently this situation requires a skilled IT expert who knows all system components and, hence, is able to identify and modify all the affected parts. However, not always we have such an expert, whereas for complex systems it is a very hard and error-prone task. In this paper we introduce DaemonX -- an evolution management framework, which enables to manage evolution of complex applications efficiently and correctly. Using the idea of plug-ins, it enables to model almost any kind of a data format (currently XML, UML, ER, and BPMN). Since it preserves relationships among the modeled constructs, it naturally supports propagation of changes to all related affected parts. We describe the general proposal of the framework and, then, its architecture and implementation.

Strigil: A Framework for Data Extraction in Semi-Structured Web Documents

In this paper we introduce Strigil, a framework for automated data extraction. It represents an easily configurable tool that enables one to retrieve a data from textual or weak-structured documents. The paper contains description of the framework architecture and its important components. Additionally, we propose a scraping language inspired by the XSL transformations designed to extract data from different kinds of documents. Although there are many different approaches focused on various aspects of data scraping, they are usually very specialized to a concrete domain or a data source. We compare these solutions and discuss their advantages and disadvantages. Our scraping language is designed to work with an ontology to map scraped data directly to classes and attributes.

Advanced REST API Management and Evolution Using MDA

The REST (Representational State Transfer) has become a popular and preferred way of communication on the Web. JSON (JavaScript Object Notation) [5] is the most used data interchange format these days. JSON can be easily used with (but not-only) applications written in JavaScript or other programming languages. In our previous work we presented a solution how to describe REST API based on the MDA principle and how to provide automatic evolution management. The aim of this paper is to extend our previous results with deeper analysis and additional model operations to make evolution process simple for the API designer. Next, we aim to enlarge abilities of the platform for the specific model of the REST to be able to generate final communication structures in JSON, or other formats like, e.g., XML, as well as their schemas. The proposed approach was implemented within a complex evolution-management framework called DaemonX which provides an extensible platform for evolution management of various data formats, such as, e.g., XML, relational, object, etc. Several experiments proving the concept were then carried out.

Adapting service interfaces when business processes evolve

Nowadays, there are many popular different notations of business process modeling, such as, e.g. BPMN [1], for a software analysis. Modeling of business processes at the conceptual level allows domain experts to cooperate in the analysis and to design the software. The whole business process model can be later translated to web services and executable BPEL [2] scripts which orchestrate all related parts together. Besides this automatization, it is necessary to define the structure of each data object in the business model. Web services usually communicate by exchanging XML documents [3]. Therefore, a software architect has to define also an XML schema of XML documents by an XML schema language, e.g. XML Schema [4]. In this paper, we present a method which derives and adapts optimal communication XML schemas for a given conceptual schema of a business process, complemented with a conceptual schema of exchanged data. We use a view on a part of a Platform Independent Model of a whole system as conceptual schema of exchanged data. We derive several XML schema documents from given inputs and choose the one which has minimal values for three metrics. The result of these three metrics is a measure of a quality of an XML schema document with respect to conceptual model of exchanged data and business rules applied on this document.

Evolution of a Relational Schema and Its Impact on SQL Queries

Since the typical feature of a real-world application is evolution, i.e. changes in user requirements, the problem of change propagation is an important task. In this paper we study its specific part – evolution of a relational database schema and its impact on related SQL queries. The proposed approach shows an ability to model database queries together with a database schema. The feature then provides a solution how to adapt database queries related to the evolved database schema. The proposal was implemented within a complex evolution framework called DaemonX and various experiments proving the concept were carried out.

Advanced Analytics of Large Connected Data Based on Similarity Modeling

Collecting various types of data about users/clients in order to improve the services and competitiveness of companies has a long history. However, these approaches are often based on classical statistical methods and an assumption of limited computational power. In this paper we introduce the vision of our applied research project targeting to the financial sector. Our main goal is to develop an automated software solution for similarity modeling over big and semi-structured graph data representing behavior of bank clients. The main aim of similarity models is to improve the decision process in risk management, marketing, security and related areas.