This issue, starting Volume 12 of the Computer Science and Information Systems journal, consists of 15 regular articles. As always, we are grateful for the hard work and enthusiasm of all authors and reviewers, which were invaluable in shaping the current issue of our journal.

The first article, “Using Part-of-Speech Tags as Deep-Syntax Indicators in Determining Short-Text Semantic Similarity” by Vuk Batanović and Dragan Bojić, describes a method for computing similarity of short texts which employs part-of-speech tags as indicators of deeper syntactic information. Unlike the majority of other tools of its kind, the presented method does not involve complex parsers or semantic role labelers, but rather relies on a statistical bag-of-words approach augmented by a part-of-speech weighting scheme. Nevertheless, the authors demonstrate competitive and superior results compared against state-of-the art methods in a paraphrase recognition task.

“Betweenness versus Linerank” by Balázs Kósa et al. provides a thorough comparison of two centrality measures for graphs and networks – betweenness and Linerank. The latter has gained popularity recently due to lower computational complexity, allowing its application on very large networks. However, the authors show that edge variants of these measures are not interchangeable, as the correlation between the two is very low. On the other hand, node variants exhibit high correlation, with betweenness still being a more accurate measure. The paper also provides clarification of Linerank computation on undirected graphs.

Darius Birvinskas et al., in “Fast DCT Algorithms for EEG Data Compression in Embedded Systems,” explore the use of the Discrete Cosine Transform (DCT) for lossy compression of electroencephalography (EEG) signals that are sent through a communication channel for further processing. The authors conclude that the method can be used in real-time embedded systems, where speed and low communication overhead are of high importance.

In their paper entitled “Developing a Fuzzy Logic Based System for Monitoring and Early Detection of Residential Fire Based on Thermistor Sensors,” Mirjana Maksimović et al. discuss a Wireless Sensor Network (WSN) approach to monitoring and determining the confidence of the occurrence of fire. Two fuzzy logic approaches with temporal characteristics are proposed in order to optimize the number of rules necessary to make correct decisions, reducing sensor activity and increasing battery life, without compromising the quality of operation.

“Distributed and Collaborative Web Change Detection System,” by Víctor M. Prieto et al., tackle the problem of keeping search-engine indexes up-to-date with the newest versions of Web pages, by detecting changes as soon as possible, with minimal computational cost. The article presents a distributed approach to solving this problem which, in an ideal scenario, would be able to detect changes in Web sites within minutes, rather than days.

Nikos Vesyropoulos and Christos K. Georgiadis, in “Customized QoS-based Mashups for the Web of Things: An Application of AHP,” consider the issue of matching Web services, in terms of selection as well as composition, with the needs of end users, based on personalized and QoS-based criteria. The authors propose a framework for solving this problem, which encompasses the multicriteria decision analysis method Analytic Hierarchy Process (AHP).

The article “Academic Data Warehouse Design Using a Hybrid Methodology” by Francesco Di Tria et al. describes the architecture of a business intelligence system for academic organizations, with data warehouse design focusing on the analysis of the factors deemed important for academic institutions, such as research evaluation and teaching. The presented approach to design is hybrid and semi-automatic, relying on ontologies in the process of integrating different data sources.

Belén Cruz Zapata et al., in their paper “Security in Cloud Computing: a Mapping Study,” offer a systematic mapping study of 344 papers in the field of cloud computing security, with the aim of assessing the current state of research and practice, and identifying future directions. The considered papers were classified and discussed according to security goal, research type and contribution type. The study concludes that cloud computing is a promising area for security research and evaluation.

In “Exploiting Geotagged Resources to Spatial Ranking by Extending HITS Algorithm,” Tuong Tri Nguyen and Jason J. Jung present a spatial ranking algorithm based on HITS, which combines and discovers meaningful relationships between spatial information and tags, allowing users to retrieve a set of relevant spatial resources with a given tag. The article evaluates the approach on a food recommendation task.

The next article, “A Software Engineering Model for the Development of Adaptation Rules and its Application in a Hinting Adaptive E-learning System” by Pedro J. Muñoz-Merino et al. introduces a software engineering model for the description of rules for adaptation of information systems. Besides rule composition, the model facilitates rule distribution and reuse, as well as collaboration among rule creators. The approach is illustrated and evaluated on an adaptive e-learning system that generates exercises with hints which can be adapted based on the rules.

Enayat Rajabi et al., in “Interlinking Educational Resources to Web of Data through IEEE LOM,” present an approach for exposing learning object metadata as linked data, based on the use of established conventions and standards. The authors focus on IEEE LOM metadata, and discusses alternative strategies for exposing links, and their tradeoffs.

“Commonalities and Differences between Requirements Engineering Tools: A Quantitative Approach,” by Juan M. Carrillo de Gea et al., compares current requirements engineering (RE) tools with the aim of answering the question of what level of variation, in terms of functionality, is observable in state-of-practice RE tools. To answer this question the authors designed a 188-item survey, which was answered by 29 participants. The results indicate the existence of three distinct clusters of tools with respect to coverage of supported features, which can help practitioners in selecting the tool most suitable to their specific needs.

In “A Novel Registration Method for High Resolution Remote Sensing Images Based on JSEG and NMI,” Chao Wang et al. takle the difficulties in applying multi-scale analysis tools to high resolution remote sensing image registration. The article presents a new registration method combining J-segmentation (JSEG) and normalized mutual information (NMI). Besides accurate extraction of feature points, the authors show that the method produces a reasonable distribution of control point pairs.

Laksono Kurnianggoro et al., in their paper “Calibration of a 2D Laser Scanner System and Rotating Platform using a Point-Plane Constraint,” explore the use of a 2D laser scanner for constructing a low-cost 3D scanner system. The two key parameters, the rotation axis and radius, are determined through careful calibration. Simulations involving synthetic data reveal low errors of the proposed method in terms of orientation degree and radius.

And finally, Laura C. Rodriguez-Martinez et al., in “Identifying Common Activities in the Graphical User Interface Development Process and their integration into the Software-System Development Life Cycle,” propose a generic model for the GUI development process (GDP), based on an identified set of activities. The article identifies common activities of previous GDPs, defines an approach to a generic GDP limited to its phases and activities, and discussed the integration of such a generic GDP with software system development life cycles. The approach is compared to other methods for GDP in the field of human-computer interaction, and illustrated on an e-commerce Web application case study.

Mirjana Ivanović

Managing Editor
Miloš Radovanović