Information and Software Technologies

In this paper, we describe a model for extracting rules that describe enterprise resource planning (ERP) system upgrade process. The rules are extracted automatically by analyzing programming code from completed ERP systems upgrade projects. Later those rules are verified and tuned by experienced programmer. The rules that we are extracting are described in the language equivalent to the first-order logic (i.e. expressivity of Turing machine). Nevertheless, we put a constrain on the rules description language by defining how knowledge base is used to define these rules. We require that ERP system code and ERP upgrade knowledge base must be transformed to a series of aligned strings without lost of expressivity. Such strong requirement ensures that we are able to use existing machine learning algorithms in the process of software development and upgrade. These series of strings are compared by strings manipulation algorithms and then differences are resolved by merge algorithm presented in this paper. We used Microsoft Dynamics NAV as an example to test usefulness of presented method but other systems can be used as well if they can be presented as series of code strings.

The problem of personal scheduling in hospitals is an important problem of applied combinatorial optimization. Numerous specialized heuristics as well as metaheuristic methods have been proposed for solving that problem. The recently proposed shift sequence based method is one of the most efficient. In the present paper we propose an enhancement making that method faster, and improving its general performance. The original and enhanced versions of the algorithm have been compared using the real data from one of the main Lithuanian hospitals.

Genetic algorithms (GAs) are a modern class of the metaheuristic methods that have been applied for the solution of different combinatorial optimization problems, among them, the quadratic assignment problem (QAP). Various modifications and alternatives of the genetic algorithms have been proposed in the artificial intelligence literature, and in many cases they performed quite successfully and efficiently. In this paper, we describe four variants of the genetic algorithm and some more or less slight variations of these variants. We also present the results of the computational experiments with both the real life like and randomly generated instances from the library of the benchmark QAP instances – QAPLIB.

Record segmentation is a core problem in structured web data extraction. In this paper we present a novel technique that segments structured web data into individual data records that come from underlying database. Proposed technique exploits visual as well as structural features of web page elements to group them into semantically similar clusters. Resulting clusters reflect the page structure and are used to segment data records. During the segmentation process the technique also generates Xpath expressions. These expressions can be later used to directly extract data records from same template generated web pages without need to redo all the clustering and segmentation processes. Extracted structured data can be reused in wide range of applications, such as price comparison portals, meta-searching, knowledge bases and etc. The experimental evaluation results of proposed technique system on three publicly available benchmark data sets demonstrate nearly perfect results in terms of precision and recall.

The amount of information that is created, used or stored is growing exponentially and types of data sources are diverse. Most of it is available as an unstructured text. Moreover, considerable part of it is available on-line, usually accessible as Internet resources. It is too expensive or even impossible for humans to analyze all the resources for a required information. Classical Information Technology techniques are not sufficient to process such amounts of information and render it in a form convenient for further analysis. Information Retrieval (IR) and Natural Language Processing (NLP) provide a number of instruments for information analysis and retrieval. In this paper we present a combined application of NLP and IR for Lithuanian media analysis. We demonstrate that a combination of IR and NLP tools with appropriate changes can be successfully applied to Lithuanian media texts.

One of the most important improvements for the Traveling Salesman Problem (TSP) heuristics is usage of limited size candidate sets, also known as candidate lists (CLs). CLs help to effectively reduce the search space and appropriately sorted biases the search in a more gainful direction. Besides the best known nearest-neighbor, quadrant, Delaunay or leading Helsgaun’s alpha CLs, other effective CLs can be constructed. We revise tour merging technique proposed by Applegate et al. and using multi-random-start procedure with incorporated fast 3-opt and simplified Lin-Kernighan-Helsgaun (LKH) algorithm modifications, generate tour union CLs and analyze them by comparing with the most common alternatives. In addition, we propose a criteria indicating when merging procedure should be terminated to form high quality CLs. Finally, we show a high potential of such CLs by performing the experiments with our modified LKH variant.

Each organization should make progress to remain competent in its business. Enterprise Modeling (EM) helps in understanding the current and planning the future states, followed by proposing improvement actions in an enterprise. To receive support from EM, we should start a process of using an Enterprise Modeling Method (EMM), that likewise any other process needs using resources. As resources are expensive, we prefer not only gaining results, rather using resources reasonably, i.e. performing an efficient process that supports the process quality. To realize if we have an efficient EMM, we should evaluate its efficiency. In this paper we present a method for efficiency evaluation of EMMs for general case of application. This method contains efficiency criteria and questions for evaluating their fulfillment. Then it is applied to the Enterprise Knowledge Development for appraising the evaluation method. The paper ends with a number of conclusions about the evaluation method.

A problem of drawing aesthetically looking graphs is considered. We focus on graphs related to management of business processes. Vertices of a graph are visualized as rectangles (flow objects), and edges are visualized as rectangular connectors (sequence flow). A particular problem of aesthetic drawing is considered where location of vertices is fixed, and the lines representing the edges should be drawn. The latter problem is restated as a graphs oriented multi-objective combinatorial optimization problem. The generally recognized criteria of aesthetic presentation, such as the general length of lines, the number of crossings, and the number of bends, are considered as the objectives to be minimized. The attitude of the potential users of the supposed algorithms towards the relative importance of the considered criteria is elicited by a psychological experiment. The elicited information is used in the development of domain-specific multi-objective optimization algorithms. We propose for that problem a version of the metaheuristics of ant colony optimization. The efficiency is evaluated experimentally using randomized test problems of different complexity.

Two approaches for evaluation of homogeneous elasticity parameters of heterogeneous material with periodic microstructure in linear elasticity are discussed in this paper. The two – scale modeling is used. Elasticity parameters for homogeneous material are evaluated in microscale and average stresses of three-dimensional heterogeneous and two-dimensional homogeneous models are compared in mezzoscale.

Changeability is a fundamental property of software systems. Every software system must evolve at all levels of abstraction (models, architecture, source code, documentation, etc.) to meet changing user and context requirements. To assess the extent of a change, change impact analysis must be performed. In this paper, we propose a taxonomy of change aspects in feature modelling domain, and analyse changeability of feature models, a high level representation of system’s external user-visible characteristics. We propose the change impact model based on a feature dependency matrix to assess validity of feature change, to follow feature change propagation and to estimate changeability of a feature model using a Jaccard distance measure. The model is implemented using Prolog logic rules. A case study is presented.

In this paper, we propose an approach based on the combination of Meta-modelling and Graph Grammars, to transform a Durational Action Timed Automata model (DATA*) into a timed refusal regiongraph (TRRG) for creating a canonical tester and generating test cases. However, our approach allows to generate automatically a visual modelling tool for DATA*, TRRG and the canonical tester. The cost of building a visual modelling tool from scratch is prohibitive. Meta-Modelling approach is useful to deal with this problem since it allows the modelling of the formalisms themselves, by means of Graph Grammars. Models manipulations are expressed on a formal basis and in a graphical way. In our approach, the UML class diagram formalism is used as meta-formalism to offer a meta-model of DATA*, TRRG and the canonical tester. The meta-modelling tool AToM

3

is used.

The major challenge for the semiconductor industry is to design devices in short time with complex logical functionality. At the very top of the list of challenges to be solved is verification. The goal of the verification is to ensure that the design meets the logical functional requirements as defined in the logical functional specification. Verification of the devices takes 40 to 70 per cent of the total development effort for the design. The increasing complexity of hardware designs raises the need for the development of new techniques and methodologies that can provide the verification team with the means to achieve its goals quickly and with limited resources.

We present a framework that enables verification test generation using circuit model presented in VHDL hardware description language. The framework consists of several tools. We use the third party tool VSYML for VHDL description translation. Our tool then extracts finite state machine (FSM) model from translated VHDL description. The goal of the model is to use it for verification test generation. We introduce our software tool and algorithm, as well. The experimental results are presented for the benchmark suite ITC’99. The obtained results demonstrated the viability of the framework.

Segmentation task can provide important information about the nature of the sequence that is understandable to humans. A new algorithm for sequence segmentation is proposed in this paper. It is shown that it is possible to segment sequence finding a nearest algebraic progression to an each segment of a given sequence. The proposed segmentation technique based of the concept of the rank of a sequence. The rank of a sequence describes exact algebraic relationships between elements of the sequence. Numerical experiments with an artificially generated numerical sequence are used to illustrate the functionality of the proposed algorithm.

There are multiple frameworks proposed conceptually defining the business and IT alignment. There are also multiple empirical methods for modeling the business and IT alignment. However, the transition between conceptual definitions and modeling languages still lacks a clear approach and tools for verifying if business and IT are aligned in a particular enterprise model. The paper presents an approach for functional business and IT alignment focusing on its implementation with nowadays enterprise modeling techniques and tools. The suggested approach is based on the SOA, GRAAL, and enterprise modeling techniques such as TOGAF, DoDAF, and UPDM. A real world example is presented to validate the suitability of the approach.

We introduced a Web Information System (WIS) adaptation architecture that is based on Software as a Service (SaaS) ideas. It includes adaptation components, which allow adaptation in two levels: the organizations and the users get their own adapted instance of the WIS. The user interface in case of multi-tenancy should be dynamically adapted according to the particular organization and user. The same application component that contains a set of fields, controls and other interface elements should be varied according to the usage context. In this paper we provide a method for adapting the user interface within the proposed adaptation architecture that uses a set of rules describing the sequence of allowed actions generated as a response to the performed user’s activities.

This paper describes adoption of Domain Driven Design and Feature Driven Development paradigms for decision support systems, using credit risk evaluation DSS as a case study. Possible development scenarios using both of these methodologies are discussed, using transformations from previously described development framework. It is concluded that these techniques might be adopted for development of complex DSS.

UML is a standard modeling language that provides several concepts which are used for different levels of design. The communication diagram is one of UML diagrams used to represent interactions between objects. But it remains an object-oriented model who owns several limits namely the duplication, the difficult resolution and reuse of models. The Aspect Oriented Modeling has shown its usefulness in the design and development of complex systems. There are several studies on the aspects composition. We propose in this paper, an approach for composition aspects. This method is inspired by the approach MATA, graph transformation to automatically integrate aspects models and Object Oriented Communication diagrams and get Aspect Oriented Communication diagrams. To achieve this transformation automatically, we propose two meta-models and from these latter’s, we propose a graph grammar. In order to validate our model transformation, we use the meta-modeling tool AToM3. A case study is presented to illustrate our approach.

The paper analyzes transformation of a correct heterogeneous meta-program into 2-stage meta-programs. We propose a methodology and describe experiments to solve two tasks: 1) transformation of the 1-stage meta-program into the set of 2-stage meta-programs; 2) checking hypothesis of the transformation equivalence under given transformation rules and constraints. The experimental results have shown that introduced formalism, rules and models ensure correctness of transformations, extend reuse dimension to automatically adapt (through transformations) variants of programs/meta-programs to different contexts of use, enable to better understand meta-program development/change processes and heterogeneous meta-programming perse.

Business Network Management (BNM) provides companies with techniques for managing their trading partner networks by making (technical) integration, business and social aspects visible within a network view and set them into context to each other. Therefore it computationally links data into business and integration networks as well as computes semantic correlation between entities of both perspectives. The linked real-world data is then captured in a network-centric variant of Business Process Modeling Notation (BPMN), which we call Network Integration Model (NIM).

In this paper, we propose an approach, which features access to the complex inter-connected business and technical perspectives in NIM, called Business Graph Protocol. We define a powerful Resource Graph (RG), which is directly computed from the underlying domain model, e.g. NIM, and allows simple, uniform, but expressive queries and traversal on the linked data. Through embedded, ad-hoc querying techniques, entry points to the network are computed. We present an approach on applying state of the art RESTful WebServices to our domain and report on our experiences with it.

Current domain analysis methods for product line engineering usually focus on the implementation of functional requirements for product lines while neglecting the quality aspects. However, in modern software system development the non-functional requirements, such as dependability, maintainability and, especially, quality, have become more important. Furthermore, quality is one of essentials dimensions of variability and there are complex direct and indirect relationships between functional and quality features of systems. The aim of this paper is to extend feature modeling for modeling software quality using elements of preference logic to aid the designer in the decision making process when selecting alternative (optional) features from a feature diagram. Preference logic is used to compactly represent and reason about preference relations between features and quality aspects in integrated feature-quality models represented using extended feature diagrams.

Testing is a very expensive but still very important stage of software development. Most applications today are designed to be networked. Often issues arise in such systems due to the interleaving of the clients and servers as well as resource racing issues. To make testing of distributed systems easier we propose an algorithm that is able to simulate few clients communicating with remote servers and is able to detect resource racing related failures. For clients simulation we use Java Path Finder (JPF). We propose JPF extension which is able to simulate and cover all possible paths threw the distributed system while two clients are interacting with the system at the same time. For resource racing problems we propose to use and extend UML class diagram with 3 additional method stereotypes. From the extended model we generate additional code constraints which are later used in testing as test oracles.

Dynamic compositional symbolic execution is a well-known white-box method for generating unit test data. However there exists certain issues when applying this method for software that heavily depends on floating point data types. In this paper we present these issues and suggest our solution. Our presented method would integrate two techniques: symbolic execution and search-based testing to increase code coverage of software under test. We have implemented it as an extension to tool PEX, that is being developed at Microsoft Research. Our extension implements search-based testing as an optimization technique using AVM method. We present coverage comparison for several benchmark functions.

A distributed service oriented system which is developed to provide physical system or process monitoring, specification, analysis and simulation services are introduced in this paper. These services in separate components are developed to be used along with computer aided design systems to provide a support for designers in designing physical systems with efficient power consumption. All components of distributed system are presented to describe their role and interaction with Energy Simulator, which is developed by authors of this paper as a key component for simulation of the physical system and for calculation of energy consumption. Then, structure of Energy Simulator and its development process as well as solutions made to create flexible and scalable service oriented architecture is presented. Finally, the implementation of Energy Simulator and a case study devoted for designing of compressed air system is described together with discussion of various constraints of service oriented performance and practical simulation applying MATLAB based tools.

This paper describes our efforts developing the voice controlled interface for the medical- pharmaceutical information system. It is well known that voice controlled interfaces are of particular importance and may provide significant convenience for healthcare professionals. Many international IT companies provide speech processing by using tools oriented at the physicians or at other professionals working in the related areas. Lithuanian professionals feel a significant lack of speech processing since there are still no attempts to integrate Lithuanian speech processing into medical information systems in this country. The paper presents several IS under development for the use in Lithuanian healthcare institutions. The experimental evaluation shows that it is possible to achieve recognition accuracy (at least 95% of correct recognition) acceptable for the practitioners. The evaluation of system prototypes shows that voice interfaces may lead to the increased convenience of IT systems at the healthcare practitioners.

The paper describes the design of a Neural Interface Based (NIS) system for control of external robotic devices. The system is being implemented using the principles of component-based reuse and combines modules for data acquisition, data processing, training, classification, direct and the NIS-based control as well as evaluation and graphical representation of results. The system uses the OCZ Neural Impulse Actuator to acquire the data for control of Arduino 4WD and Lynxmotion 5LA Robotic Arm devices. The paper describes the implementation of the system’s components as well as presents the results of experiments.

Influence of network communications to the final performance of grid visualization software VisPartDEM is investigated. Distributed architecture of VisPartDEM is designed for interactive visualization of large particle systems simulated by the discrete element method on the gLite-based grid. Performance analysis of VisPartDEM is made on the geographically distributed gLite grid infrastructure. The attention is focused on the data transfer from storage elements, image transfer to clients and their contribution to the total benchmark time. The discussed issues are important for all researchers interactively visualizing results on gLite-based remote grid infrastructures.

Information and communication technologies are an integral part of modern society. The need of highly qualified professionals with a master’s degree is growing. Universities are faced with major challenges in organizing postgraduate studies. For a variety of reasons, very high percentage of postgraduate studies isn’t completed. The obvious forces the need for innovative forms of education. One of the ways to solve problems in postgraduate studies are using of blended learning.

Generative Learning Objects (GLOs) introduced by Boyle

et al.

in 2004 extend the reuse dimension and create conditions for better quality in learning and teaching. We motivate the benefits of GLOs for e-learning and extend the scope of understanding GLOs by introducing the concept of the two-stage GLOs (TS GLOs). The aim of the paper is to show how this new kind of GLOs can contribute to the advanced e-learning through e-learning variability (social context, content variants, e-learning environment characteristics, etc.) modelling. GLOs and TS GLOs are seen as tools to further enhance reuse, to ensure a great deal of flexibility for teachers, and to manage complexity in the LO domain. Our case study has also showed the benefit for students in using the approach for self-learning by selecting different tasks or variants with respect to students flavour or abilities.

The growth in the use of mobile devices in everyday life determined an increasing demand to access educational information using mobile technology. Nevertheless, most existing computer based Learning Management Systems still do not have advanced access support for mobile devices. This paper presents the basis of an in-progress research, aiming to enhance our understanding of mobile learning usability considerations and measurement. It also provides the starting point for performing a usability evaluation of the MyMobile web interface of our university’s Moodle LMS. The main contribution of this paper is our proposed approach, based on four perspectives that have to be considered when designing or testing a mobile learning solution: Pedagogical usability, Usability of the device, Usability of the content and Usability of the mobile web interface. Additionally, metrics, methods and guidelines for usability testing are given. The main goal of the paper is to provide an overview on a proposed framework for testing and optimizing a LMS mobile web interface from a usability perspective.

Interdisciplinarity will become, not only a way of content restructuring, but also a new way to organize learning. This paper proposes a framework for building ontology for e-learning and aims for a navigation and efficient retrieval and reuse content in the “virtual university”. The framework used the content annotation Learning Object (LO), and the content structure by exploiting Learning Object Metadata (LOM). Ontology can be used for content analysis, recognition of concepts, the relationship between them and allows search and retrieval of content produced by query and the reuse of it in different subjects, using the MP7QF framework specific to MPEG-7.

The use of information communication technologies (ICT) in education has introduced a new set of technological and educational opportunities for educators and students. That is why it is so important to understand the attitudes of educators towards e-learning in education as well as ICT based learning.

Information and Communication Technologies are important enablers of the new social structure. We are experiencing the first generation of truly portable ICT removable memory chips, diaries, email, Web, basic word processing and spreadsheets, and data input, storage, and transfer what is used in nowadays education process. The aim of the papers is to present the new architecture based on the open source technological solutions to be used in e-learning and social networking systems.

In this paper we introduce a framework for providing competition based social learning. We discuss that competition in learning activities, especially paired with social learning, is very well accepted still the existing solutions lack tools for directing and organizing learning activities. We provide a description of a social learning portal www.arzinai.lt which implements competition in learning, also has features of personal and group learning environments. Some usage statistics, which confirm that competitions provide a strong motivation for learners, are provided as well.

Implementation of agent based intelligent technologies in the environment of distance learning can individualise learning, enable an optimal use of learning resources and ensure effective interaction with all participants of the learning process. In recent scientific literature, however, only descriptions of individual agents or small agent groups that support one learning activity may be found. The article describes a designed system of pedagogical software agents that integrally supports the learning process and that is adapted to virtual learning environment Moodle.

Topic Maps is knowledge representation and exchange model having rich verbalization possibilities for improving navigation-based search and understanding for human and interpretable by machine. Despite increasing usage of Topic Maps in knowledge-based information networks and portals, the flexibility of the representation often leads to disability to judge on correctness or completeness of a Topic Map model, or simply gives no rule how to model problematic actualities. The paper presents the methodology for developing Topic Maps on the base of the principles of developing ontologies extending them with recommendations for solving some problems specific to Topic Maps development: creating normalized schema; representing association properties and names, multidimensional classification hierarchies, and so on. The suitability of the methodology is analyzed via a case study for developing a topic map of Information System Development Methodologies with Ontopia tools.

Structured language, based on Semantics of Business Vocabulary and Business Rules (SBVR), can be seen as domain expert friendly means for developing OWL2 ontologies, which are becoming more and more important in Semantic Web and Enterprise applications. The goal of the paper is to present transformations from SBVR specifications to ontologies and to describe conditions for creating “right” vocabularies in order to obtain consistent ontologies without losing information. The need for such approach is caused by several reasons. Concept models rely on the closed world assumption, whereas ontologies rely on the open one where every constraint should be explicitly specified. Both SBVR and OWL2 have terminology related part, desirable being separated from the substantial ontology. We suggest rules that can help creating meaningful SBVR vocabularies regarding consequences of affecting the behavior of ontology reasoners, taking advantages of ontologies and retaining terminological information separately from the main ontology.

The goal of the paper is to present question patterns in structured natural language and their transformations into ontology query language SPARQL for allowing business participants to communicate with business software services and data in more flexible and friendly way. The structured language is based on Semantics of Business Vocabulary and Business Rules (SBVR) metamodel, which allows creating and managing business vocabularies and business rules in specific domains. The current paper is focused on transforming question patterns, including usage of synonyms and synonymous forms; projecting formulations constrained by atomic formulations based on facts and fact types; projections on several variables; restricting query results by auxiliary variables constrained by various logical formulations; supplementing questions with derivation rules from SBVR vocabulary of business rules, etc. Patterns are followed by examples, proved by implemented SBVR query editor and SBVR to SPARQL transformations.

In this study, test case prioritization approaches that are used to execute the regression testing in a cost-effective manner were investigated. We discussed the critical issues and best practices that a software company should focus on before and after the implementation of test case prioritization techniques inside the company. Due to the increasing complexity of today’s software intensive systems, the number of test cases in a software development project increases for an effective validation & verification process and the time allocated to execute the regression tests decreases because of the marketing pressures. For this reason, it is very crucial to plan and setup test case prioritization infrastructures properly in software companies to improve the software testing process. Ten best practices for a successful test case prioritization are introduced and explained in this study.

Even though the software industry seems to have matured from the initial stage, the software projects’ success rates are still low. This is mainly because of lack of correct, unambiguous, complete, consistent description on software requirements. How to specify and represent requirements correctly, unambiguously, completely and reach a common understanding among stakeholders in software projects, have become high priority issues. The aim of this paper is to study the ways of specifying and representing the semantic information of functional requirements in software projects. We propose a meta-model and implement it into semantic forms in Semantic MediaWiki. Our approach enables the building of functional requirements on a common semantic basis, thereby improving the analyzability and consistency of the requirements. The wiki environment also enables asynchronous collaboration to create and maintain the functional requirements.

Software requirements are the concurrent part of any project. However, these requirements are not effectively used in the testing process. The solutions founded in literature not only confirm the relevance of the software requirements and testing, but also reveal the absence of better solution combining software requirements and testing in a single element. The solution how to minimize the gap between requirements and testing is presented in this paper, since it is especially significant in a change of requirements. The idea of solution is based on gathering the useful data for information system testing from requirements and transforming it into structural models.

As software systems evolve, it becomes increasingly complex for maintainers to keep them aligned with rapidly changing business requirements. Therefore the cost of software maintenance often exceeds the cost of its initial development or adaptation. As a result, automated approaches for software comprehension emerge providing valuable improvements and cost-savings for the software maintenance. This paper presents an approach that facilitates software comprehension by enabling traceability of implementation of business rules and business scenarios in the software system. It also describes a case study on application of this approach for comprehension of business logic implemented in the enterprise content management system and reports obtained results.