Informatica Economica Vol. 15 No. 4/2011

Issue Topic: Technologies for the Semantic Web and Software Agents

Issue Editor: Catalin BOJA, PhD


Grigore C. Moisil Award 2011 5

University 4.4 – A Development Strategy for Education and Research Centers

Catalin BOJA, Cristian TOMA 7
Due to increased demand for qualified human resources, for 'new & rare skills', for software solutions, reliable products and services in the field of applied informatics, there are large available financial funds that can be accessed by Informatics and Cybernetics schools. Edu-cational and research departments must capitalize funds provided by the Europe-an/international institutions and private companies, by supporting the creation of spin-off en-tities that will conduct technology transfer projects. These funds must be used to increase the quality of teaching and to improve research results by assuring the financial needs and tech-nical resources of teachers (project based payments), students (scholarships projects) and the community (public available projects). The presented strategy, University 4.4 describes four development directions for a four years period. It has been developed by Catalin Boja, Razvan Bologa, Marius Popa and Cristian Toma and since November 2011 it represents the assumed development strategy of The Department of Economic Informatics and Cybernetics (DICE) from The Bucharest University of Economic Studies.
Keywords: University, Research, Strategy, Education, Informatics, Cybernetics, Department, DICE

Distorted Fingerprint Verification System
Fingerprint verification is one of the most reliable personal identification methods. Fingerprint matching is affected by non-linear distortion introduced in fingerprint impression during the image acquisition process. This non-linear deformation changes both the position and orientation of minutiae. The proposed system operates in three stages: alignment based fingerprint matching, fuzzy clustering and classifier framework. First, an enhanced input fingerprint image has been aligned with the template fingerprint image and matching score is computed. To improve the performance of the system, a fuzzy clustering based on distance and density has been used to cluster the feature set obtained from the fingerprint matcher. Finally a classifier framework has been developed and found that cost sensitive classifier produces better results. The system has been evaluated on fingerprint database and the experimental result shows that system produces a verification rate of 96%. This system plays an important role in forensic and civilian applications.
Keywords: Biometric, Fingerprints, Distortion, Fuzzy Clustering, Cost Sensitive Classifier

The Management of Manufacturing-Oriented Informatics Systems Using Efficient and Flexible Architectures
Constantin-Daniel AVRAM, Cecilia CIOLOCA, Laurentiu CIOVICA, Mihai CURTEANU, Mihai GEORGESCU 22
Industry and in particular the manufacturing-oriented sector has always been researched and innovated as a result of technological progress, diversification and differentiation among consumers' demands. A company that provides to its customers products matching perfectly their demands at competitive prices has a great advantage over its competitors. Manufacturing-oriented information systems are becoming more flexible and configurable and they require integration with the entire organization. This can be done using efficient software architectures that will allow the coexistence between commercial solutions and open source components while sharing computing resources organized in grid infrastructures and under the governance of powerful management tools.
Keywords: Manufacturing-Oriented Informatics Systems, Open Source, Software Architectures, Grid Computing, Web-Based Management Systems

SOA and BPM, a Partnership for Successful Organizations
Gheorghe MATEI 39
In order to stay effective and competitive, companies have to be able to adapt themselves to permanent market requirements, to improve constantly their business process, to act as flexible and proactive economic agents. To achieve these goals, the IT systems within the organization have to be standardized and integrated, in order to provide fast and reliable data access to users both inside and outside the company. A proper system architecture for integrating company’s IT assets is a service oriented one. A service-oriented architecture (SOA) is an IT architectural style that allows integration of the company’s business as linked, repeatable tasks called services. A subject closely related to SOA is Business Process Management (BPM), an approach that aims to improve business processes. The paper also presents some aspects of this topic, as well as the relationship between SOA and BPM. They complement each other and help companies improve their business performance.
Keywords: Information Systems, SOA, Web Services, BPM

Study on the Support Systems for Corporate Governance
Claudiu BRANDAS 55
The problems generated by the information asymmetry within the agency relationships at corporation level, governments and capital markets led to a higher necessity for corporate governance (CG). The information system of corporate governance is a very complex one, which involves a series of actors, decision-making and transactional processes, technologies, procedures and good practice codes. In order to ensure the efficiency and efficacy of corporate governance as a premise for increasing company’s performance and position consolidation for the company on the capital market, it is necessary to analyze the way information technology could contribute to this undertaking. The purpose of this study is to analyze the architecture of the corporate governance systems and to identify and classify the systems and technologies involved in ensuring CG support in order to underlie the basis for developing a conceptual model for a hybrid and collaborative support system for corporate governance.
Keywords: Corporate Governance, Support Systems, Information Technology, Decision-Making, Performance

A Survey of Wireless Fair Queuing Algorithms with Location-Dependent Channel Errors
The rapid development of wireless networks has brought more and more attention to topics related to fair allocation of resources, creation of suitable algorithms, taking into account the special characteristics of wireless environment and insurance fair access to the transmission channel, with delay bound and throughput guaranteed. Fair allocation of resources in wireless networks requires significant challenges, because of errors that occur only in these networks, such as location-dependent and bursty channel errors. In wireless networks, frequently hap-pens, because interference of radio waves, that a user experiencing bad radio conditions during a period of time, not to receive resources in that period. This paper analyzes some resource allocation algorithms for wireless networks with location dependent errors, specifying the base idea for each algorithm and the way how it works. The analyzed fair queuing algorithms differ by the way they treat the following aspects: how to select the flows which should receive additional services, how to allocate these resources, which is the proportion received by error free flows and how the flows affected by errors are compensated.
Keywords: Fair Scheduling, Wireless Networks, Location Dependent Channel Errors, Sched-uling Algorithms

A Mining Algorithm for Extracting Decision Process Data Models
Cristina-Claudia DOLEAN, Razvan PETRUŞEL 79
The paper introduces an algorithm that mines logs of user interaction with simulation software. It outputs a model that explicitly shows the data perspective of the decision process, namely the Decision Data Model (DDM). In the first part of the paper we focus on how the DDM is extracted by our mining algorithm. We introduce it as pseudo-code and, then, provide explanations and examples of how it actually works. In the second part of the paper, we use a series of small case studies to prove the robustness of the mining algorithm and how it deals with the most common patterns we found in real logs.
Keywords: Decision Process Data Model, Decision Process Mining, Decision Mining Algorithm

Testing APT Model upon a BVB Stocks’ Portfolio
Alexandra BONTAŞ, Ioan ODĂGESCU 96
Applying the Arbitrage Pricing Theory model (APT), there can be identified the major factors of influence for a BVB’ portfolio stocks’ trend. There were taken into consideration two of the APT theory models, establishing influences upon portfolio’s yield: given to macroeconomic environment and to some stochastic factors. The research’s results certify that, on the long term, what influences the stocks’ movement in the stock market is mostly the action of specific short-term factors, without general covering, like the ones that are classified in the research area of behavioral finance (investors’ preference towards risk and towards time).
Keywords: Portfolio, Risk, Stocks, Yield, Testing

A Perspective on the Benefits of Data Virtualization Technology
Ana-Ramona BOLOGA, Razvan BOLOGA 110
Providing a unified enterprise-wide data platform that feeds into consumer applications and meets the integration, analysis and reporting users’ requirements is a wish that often involves significant time and resource consumption. As an alternative to developing a data warehouse for physical integration of enterprise data, the article presents data virtualization technology. There are presented its main strengths, weaknesses and ways of combining it with classical data integration technologies. Current trends in data virtualization market reveal the great potential of growth of these solutions, which appear to have found a stable place in the portfolio of data integration tools in companies.
Keywords: Data virtualization, Data service, Information-as-a-service, Integration middleware

Framework for Evaluation of the IT&C Audit Metrics Impact
Marius POPA 119
The paper defines an assessment system for performance of IT&C audit process. The analytical models of performance indicators are provided together with the interpretation of their results. Performance levels catch the quality characteristics of the audit processes carried out for distributed informatics systems. Also, the paper presents a performance assessment framework for audit processes and a performance audit methodology. The impact of performance indicators is defined as the organization’s income after performance audit recommendation implementing. Methods and techniques for performance assessment are provided for audit processes of the distributed informatics system. The impact levels of performance indicators are calculated before implementation of the performance recommendation and after that to establish whether the performance audit increases the quality of IT&C audit processes.
Keywords: Performance Metrics, Metric Impact, Audit Process

RAD Applied in the Context of Investment Banking Trading Systems Development
Iosif ZIMAN 134
RAD as a methodology for implementing information systems has been used in a broad range of domains utilizing technology as an informational backbone but perhaps one of the main areas where this approach has been proven to be a natural fit has been in the investment banking (IB) industry, most notably when applied to trading systems. This paper introduces some of the main tenants of RAD development and focuses on a number of case studies where RAD has proven to be an extremely suitable method for implementing solutions required in the IB industry as well as explaining why RAD may be more successful than other classic de-velopment methods when applied to IB related solutions.
Keywords: RAD, Information Systems, Investment Banking, Trading Systems

Social Transparency through Recommendation Engines and its Challenges: Looking Beyond Privacy
Remus TITIRIGA 147
Our knowledge society is quickly becoming a ‘transparent’ one. This transparency is acquired, among other means, by ’personalization’ or ‘profiling’: ICT tools gathering contextualized information about individuals in men–computers interactions. The paper begins with an overview of these ICT tools (behavioral targeting, recommendation engines, ‘personalization’ through social networking). Based on these developments the analysis focus a case study of developments in social network (Facebook) and the trade-offs between ‘personalization’ and privacy constrains. A deeper analysis will reveal unexpected challenges and the need to overcome the privacy paradigm. Finally a draft of possible normative solutions will be depicted, grounded in new forms of individual rights.
Keywords: Recommendation Engines, Profiling, Privacy, ‘Sui Generis’ Copyright

The BWS Open Business Enterprise System Architecture
Cristian IONITĂ 155
Business process management systems play a central role in supporting the business operations of medium and large organizations. This paper analyses the properties current business enterprise systems and proposes a new application type called Open Business Enterprise Sys-tem. A new open system architecture called Business Workflow System is proposed. This ar-chitecture combines the instruments for flexible data management, business process manage-ment and integration into a flexible system able to manage modern business operations. The architecture was validated by implementing it into the DocuMentor platform used by major companies in Romania and US. These implementations offered the necessary data to create and refine an enterprise integration methodology called DM-CPI. The final section of the paper presents the concepts, stages and techniques employed by the methodology.
Keywords: BWL, Workflow, BWS, Evaluation, Open Business Enterprise System, DM-CPI

Reliability in Distributed Software Applications
Cătălin Alexandru TĂNASIE, Sorin VÎNTURIŞ, Adrian GRIGORIVICI 167
Reliability is of vital importance for distributed software application and should be ensured in all stages of the development cycle. Ensuring a high level of reliability for distributed software applications leads to competitive applications which increase the level of user satisfaction. The aim of this paper is to present techniques and methods which ensure high level of reliability. A model for estimating the reliability through risk assessment is presented. Distributed software applications are composed of multiple components spread across multiple heterogeneous platforms and partial failures are inherent. To ensure high reliability is very important that the input data for distributed application components are correct and complete.
Keywords: Distributed Applications, Reliability Model, Risk Assessment, Data Acquisition

Clustering Analysis within Text Classification Techniques
Madalina ZURINI, Catalin SBORA 178
The paper represents a personal approach upon the main applications of classification which are presented in the area of knowledge based society by means of methods and techniques widely spread in the literature. Text classification is underlined in chapter two where the main techniques used are described, along with an integrated taxonomy. The transition is made through the concept of spatial representation. Having the elementary elements of geometry and the artificial intelligence analysis, spatial representation models are presented. Using a parallel approach, spatial dimension is introduced in the process of classification. The main clustering methods are described in an aggregated taxonomy. For an example, spam and ham words are clustered and spatial represented, when the concepts of spam, ham and common and linkage word are presented and explained in the xOy space representation.
Keywords: Knowledge Societies, Text Classification, Spatial Representation, Artificial Intelligence, Clustering Analysis, Spam Filtering

A Decision Support System (DSS) for Project Management in the Biodiesel Industry
Alin Paul OLTEANU 189
The European biodiesel industry is currently facing several challenges affecting the profitability of investment projects in the industry. Among these challenges are higher prices for oilseeds, which are the main input for biodiesel production, lower fiscal support by national governments for biodiesel producers and high price volatility of oil markets. Thus identifying all opportunities for optimizing the value chain and lower the production cost of biodiesel is a main requirement for an efficient project management in the biodiesel industry. The paper addresses this topic by developing a decision support system tailored to the needs of Romanian investors in biodiesel production. The system optimizes the main activities of the biodiesel value chain and supports the decision making process at management level. In addition the DSS enables the user to perform sensitivity analysis based on varying various input parameters.
Keywords: Biodiesel, Investments, DSS, Optimization, Management

Designing and Implementing a Distributed Social Network Service for Mobile Devices
Alexandru RADOVICI, Valentin CRISTEA 203
The paper presents a new paradigm for building social network services. The proposed platform is called eXtensible Social Network. As it uses the XMPP protocol for authentication and communication, it allows users from different service providers interact with each other, without having to change their accounts. Moreover, the platform provides means for users to authenticate and interact with each other while temporary disconnected from the Internet. Moreover, the platform is specially designed for mobile devices, running on their restrictive operating systems and taking advantage of the systems optimizations.
Keywords: Social Network, Mobile Devices, Service, XMPP, iOS, Android, Distributed, Platform

Call for Papers 216

The Eleventh International Conference on Informatics in Economy, IE 2012

The 5th International Conference on Security for Information Technology and Communications, SECITC’12

Publishing Guide for Authors 218

Inforec Association 220