Revista Romana de Interactiune Om-Calculator

111-126

Authors: Marian Cristian Mihaescu

This paper presents a study on a design methodology for the interface of a recommender system. A usability evaluation is designed for an intelligent interface of a recommender system that runs along Tesys e-Learning platform. Building intelligent e-Learning systems enables students to access the educational resources is such a way that the learning effort is optimal when a high degree of knowledge is required. The design of the interface follows an iterative process in which low fidelity prototypes are built and evaluated. The usability evaluation is performed in two ways. A questionnaire is proposed for evaluation of low fidelity models. For high quality results a data analysis procedure of task completion times is used.

141-154

Authors: Costin Pribeanu

In recent years the educational applications based on augmented reality (AR) gained interest due to the opportunities provided by the integration of real objects from the traditional didactics on a hardware-software platform. The implementation of the AR concept, which is based on the superposition of computer generated images over the observed image of a real object on the screen, may lead to specific usability problems that are reflected on the ease of learning how to operate, ease of use, and comfort in use. In this paper a model with causal indicators is presented that is influencing the ergonomic quality of an AR-based educational application. The results show that the ease of reading the information on the screen and the ease of selecting a chemical element are the most important causal indicators for the ergonomic quality of the application.

169-192

Authors: Adrian-Nicolae Zamfirescu, Traian Eugen Rebedea

At the border of natural language processing and information retrieval, named entity recognition has represented one of the most important research problems of the two domains, that has not been solved perfectly yet even for English texts. Furthermore, named entity recognition has opened up the path of solving other problems that use these linguistic contructs, such as the identification of quotes and declarations made by persons, in general, but also by companies or other types of organizations, or the extraction of events from texts. The problem of named entity identification and clasification has appeared from the necessity of being able to report the appearances of names of persons, organizations and other types of named entities relevant for various domains within written documents. In this article we shall present a solution for solving these three aforementioned problems for texts written in Romanian from various sources, like news items, blog articles or comments from social newtorks. The paper starts with a short overview of the theoretical underpinnings used for solving these problems, then we will present the methods actually used for the designed solution for Romanian. It combines machine learning algorithms with heuristics based on text patterns and regular expressions. At the end, we shall highlight the accuracy of the various methods used for solving the tasks, together with a comparison between the results obtained by each method.

1-22

Authors: Vasile Danut Mihon, Vlad Colceriu, Victor Bâcu, Dorian Gorgan

The analysis of large volumes of data, the extraction of relevant information, and the interactive exposure and visualization of the results represent the basic processes in modeling the natural phenomena, based on satellite image processing. Due to the large dimension and their high frequency of acquisition, the satellite images are stored within specialized data repositories and accessed by means of standards. The paper presents the experiments through the GreenLand platform to achieve high usability user interaction techniques for accessing, searching, extracting, and visualizing the satellite images. The implemented solutions use the geospatial services provided by the OGC standard. The benefits resulted by integrating this standard with the GreenLand platform and the services given by the Grid infrastructure represent another main subject of this paper.

35-56

Authors: Alin Moldoveanu, Anca Morar, Victor Asavei

Mixing concepts and techniques coming from 3D MMO games, virtual reality, mixed reality and ubiquitous computing, the 3DUPB project challenges the traditional ways of our daily activities and interaction. În a few words, as a computer system, it can be described as "3D MMO Virtual and Real-Time Replica of the POLITEHNICA University of Bucharest campus". However, tightly integrated with the educaţional, social and organizaţional life of the university, 3DUPB will be more than just a computer system and will open new possibilities for human activity and interaction, a glimpse at how mixed reality systems can shape the future. This paper describes the main reasons while 3D UPB was started, its up to date concept, relation with other systems, the basic software architecture, the main technical issues and solutions, as well as the perspectives.

79-96

Authors: Marian Dardala, Titus Felix Furtuna, Adriana Reveiu

This paper presents how to design and implement a software module to achieve an alternative description of a map thematic compositions and applications of this representation. It starts from a layer containing spatial data of polygon type that can be described through adjacency matrix that holds neighborhood information for each of the existing vectors in the spatial feature. There are also presented applications of this representation: coloring maps with a minimum number of colors and the Moran index of spatial correlation calculation. The maps coloring is done thus for any two neighboring polygons they do not have the same color. The Moran spatial correlation coefficient is the standard used to determine the degree of dependence between the values of a variable in neighboring locations. It is used to identify a pattern in the spatial distribution of the valuesof an economic variables and the existence of common factors that influences the obtaining of these values.

1-14

Authors: Tiberiu Boroş, Adrian Zafiu

Transliteration has been previously introduced for translating proper names between languages, when the two languages are incompatible at the phonetic level and they use very different alphabets. We propose an data-driven method for transliteration from English to Romanian and we present a series of applications based on this task, such as text-to-speech synthesis (on multilingual text) and what we call “perception based search”.

35-58

Authors: Alexandru Balog

he purpose of this study is to test a model of acceptance of e-learning based on UTAUT model developed by Venkatesh, Morris, Davis and Davis (2003). After a brief presentation of the UTAUT model and recent studies in the field of e-learning, are described the proposed model and the hypotheses, methods used and results obtained. By applying multivariate analysis methods on a sample of students (N = 224), the results show that the model is viable in the context of e-learning in Romania. The model explained 63% of variance of Intent to continue the use of e-learning.

81-96

Authors: Cristina Serban, Irina Grosu, Alexandra Siriteanu, Alexandru Averescu, Adrian Iftene

This article presents a client-server application that, enabling the user to remotely control with an Android component the applications running on the Microsoft Windows operating system. The system consists of two main modules: the client (the application running the user’s phone) and the server (the application running on the computer). Additionally, there are two means of communication between the two key components: Wi-Fi and Bluetooth. Tests we have done on this solution demonstrated the robustness of the application in use over a long period of time, stress testing and communication when there are objects between client and server.

1-12

Authors: Peter Forbrig

One challenge in software engineering is the development of smart environments that help users to intuitively accomplish their tasks. The ideal smart environment dynamically manages a diverse collection of devices, is accessible by multiple users and effectively supports the users’ tasks. The design of smart environments relies on detailed models of devices, users and their tasks. In this paper, we present our modelling language “CTML” specifically developed for smart environments. We demonstrate how the language was designed and how it was used for usability evaluations in a virtual smart environment. We then discuss the importance of “task migrateability” – a usability principle often neglected by contemporary smart environments. We argue that the proper implementation of this usability criterion can improve the usability of smart environments. Finally, we investigate how tangible user interfaces are related to smart environments and how this interaction technique can be used to support task migrateability.

19-25

Authors: Dragos Daniel Iordache, Ioan Neacsu

In this paper are presented the evaluation results with teachers of added value by economic and psychological adequate utilization of a learning-teaching platform based on augmented reality technology. Two didactical applications developed on this platform in the framework of international project ARiSE have been evaluated. The topic of the first application is biology and the second has been developed for learning chemistry knowledge (atom structure, periodic table). We used the focus group as evaluation method, with two interview guides related to each application.

31-36

Authors: Andreea Chiscari, Dorian Gorgan

Sketching is often used by people to express ideas. Some concepts that are hard to explain in words can be easily expressed using a figure or drawing. As the pen-based user interfaces became common, many systems that use this type of interaction were built. The traditional methods used in sketch recognition can be grouped into three categories: aspect-based recognition, gesture-based recognition and geometry-based recognition. Although each of these methods has advantages and disadvantages, the geometry-based methods have proved to be the most suitable for building multi-domain recognizers. As an extension to the existing eLearning systems, we propose the integration of a sketch recognition system that can be used in many educational areas. The recognition system supports the user in the drawing process and offers meaningful and relevant information about the recognized concepts. In order to use questions that require a graphical answer, before the system can correctly and automatically evaluate the student's answer and offer immediate feedback, the teacher must provide a reference solution to the problem, the correct/expected answer. The system uses its knowledge, consisting of primitive forms and constraints) to recognize the graphical primitives and the relations between them and compares the student’s solution with the one given by the teacher.

41-44

Authors: Cristina Serban, Irina Grosu, Alexandra Siriteanu, Alexandru Averescu, Adrian Iftene

This article presents an application that the authors have developed for the Android platform, which allows a user to remotely control the applications on a computer which has the operating system Microsoft Windows. There are many other applications that have similar functionalities, such as Vectir, PPT Remote, Remote presenter for Android or Tango Remote Control Media Player HD, to name a few. Our application is comprised of two main modules: the client (the application running on the user's smartphone) and the server (the application running on the computer). In addition to these, there are two means of communication between the two modules, based on Bluetooth and Wi-Fi. The Android application (the client) is able to connect with any PC running a server background application. This component offers the user two screens: one that lists the available desktops and the means of communication (Bluetooth or Wi-Fi) and one that lists the applications that the user can control together with the actions available for each of them. The Desktop component (the server) contains three main modules. The Accessible Applications module keeps a list of the applications that can be accessed by the application. It gets this information from an .xml configuration file. The Running Applications module monitors the running processes in the computer. It is its role to notify the client whenever an application is started or exited. By receiving commands from the Android Component, we control applications like Microsoft Power Point and Windows Media Player. The communication consists of passing strings with a certain format between the client (the Android Component) and the server (the Desktop component). The application has been evaluated based on relevant use cases and tests, such as functionality tests, distance tests, stress tests.

49-54

Authors: Arnold-Mihai Waldraf, Dorian Gorgan

This paper presents the development and the experiments on a Web services based eLearning platform, called eLaWS. The authors describe the concepts and the techniques that support the flexible and dynamical development of educational materials and the improvement of functionality by using a set of Web services. The new functionality is dynamically added by the eLearning platform. The user model, dynamic content lesson models, and communication mechanisms are highlighted and detailed throughout the paper. A few experimental use cases allow the evaluation and the validation of the eLaWS platform.

65-70

Authors: Teodor Paul Alin Becan, Dorian Gorgan

This paper aims to challenge the problem of finding accurate and relevant search algorithms in order to obtain the best audio output in terms of intelligibility and naturalness, the usually employed measures to describe the quality of synthetic speech. It presents the approach used for searching the sound components that correspond to a provided phonetic transcription input of the Romanian language in a given acoustic database. We integrate this search module into the Unit Selection component of a text-to-speech system for Romanian language.

77-80

Authors: Vlad Posea

The article describes the first steps towards a semantic competence management system. The system is designed to improve the activity of a recruiting company specialized in the IT domain and it uses technologies from the domains of semantic web and web services.

85-88

Authors: Ion Smeureanu, Narcisa Isaila

The usability and accessibility are considered two major attributes in designing of necessary software for computer assisted learning through assistive technology (for disabled). Assessed separately, most evaluations are the result of researches in usability, which are addressed to all, while for people with disabilities, especially visual, the accessibility is a priority. The strong connection between accessibility and usability is based on the needs of users in achieving some clearly defined objectives so the accessibility of software products should be considered in terms of their use by people with disabilities. The paper aims to present the result of researches conducted in two centers of education for people with visual disabilities, concerning the assessment of software for learning math, according to its accessibility and usability features.

93-96

Authors: Stefan Negru, Sabin Buraga

Personas is a one of the most used method for gathering and presenting the user preferences in the context of the Human-Computer Interaction. However, a few studies have focused on linking information obtained via the Personas method with other information involved into the product development and testing stages or creating a specific conceptual model. In this article, we consider several Semantic Web technologies such as microdata and RDFa, in order to provide a solution to these issues by proposing a conceptual model for personas (a HTML5 microdata schema).

103-108

Authors: Diana Lupan, Mihai Dascalu, Stefan Trausan-Matu, Traian Rebedea, Philippe Dessus, Maryse Bianco

The analysis of written communication can reveal subtle information, such as speaker’s emotional state, attitude and intentions. However, these cannot always be extracted accurately, at a level comparable to humans’ abilities. Extracting emotions does not only reveal information about the speaker, but can also anticipate what effect might the text have on the person who reads it. Therefore we propose an automatic method for analyzing emotions induced by texts, more specifically a reader’s most likely emotional state after reading a news article. In order to achieve this, we combined a context independent approach (analyzing the text that is read using natural language processing tehniques) with the influences of user’s current emotional state evaluated through his/her feedback.

113-116

Authors: Ana Maria Marhan

As the potential impact of Internet use on suicidal behaviour is currently under questioned, experts have yet not conclusively ruled on the extent of this problem. At the moment, no one really knows what kind of information is exposed to a disturbed individual while surfing the web, nor how difficult is for a typical Romanian Internet user to find suicide-related information. Hence, this paper aims to initiate a discussion on some basic issues related to the quality and accessibility of suicide-related information available to the Romanian-speaking user, as a prerequisite in harnessing the potential offered by Internet technologies in prevention, but also intervention in suicide-related cases.

121-124

Authors: Julia Kovacs, Dorian Gorgan

This paper presents the implementation of a face recognition method and testing techniques on mobile phones with Android operating system. The method is based on the Eigenfaces algorithm, which allows variations of face expressions. The Eigenfaces approach is often used because of its simplicity of implementation and learning ability, there is no need of specific geometry or explicit descriptions of the face. The performance of the algorithm is experimentally evaluated by calculating the correct and false acceptance rate and the correct and false rejection rate. The algorithm is implemented as a Web Service whose methods are called from an application, developed for mobile phones with Android operating system.

129-134

Authors: Danut Mihon, Vlad Colceriu, Angela Minculescu, Victor Bâcu, Dorian Gorgan

Our society needs and environmental predictions encourage the applications development, oriented on supervising and analyzing different Earth Science related phenomena. This paper experiments new methods in satellite images processing throughout the GreenLand application. Visual descriptions of these methods represent one of the main objectives of this platform, new features and concepts being defined for this purpose. GreenLand facilitates the complex workflows (called hyper-graphs) development, based on two editor tools. The gProcess and ESIP (Environment Oriented Satellite Data Processing Platform) platforms have an important role in their development, by providing services for workflow creation and management. Due to large volume of data that need to be processed, the GreenLand operators and workflows are executed over the Grid infrastructure. This platform offers high storage and computing resources, capable of applying parallel and distributed execution techniques.

1-18

Authors: Teodor Stefanut, Dorian Gorgan, Nikolas Dovrolis, Eleni Kaldoudi

The development of teaching materials has currently became one of the main concerns of the specialists from various domains, including medicine. The continuous development of e-Learning applications demands the creation of new types of resources that encourage individual study activities with automatic system monitoring and less involvement from the teacher. One of the most promising approaches used currently for creating this type of educational materials in an efficient manner is the improvement and repurposing of the already existing resources. Nevertheless, this method generates new issues related with the description of the connections between the involved resources and also with the search process and identification of the relevant, good quality materials. In this paper we present a possible solution to these problems based on the characteristics of the social networks blended with functionalities specific to the semantic web. Through these tools we will describe the relations between resources and will consequently be able to provide more significant search results. The application presented, called Metamorphosis+, exposes a user interface that hides the complexity of the description process and allows users without technical background to benefit from the provided functionalities.

37-54

Authors: Paul Fogarassy-Neszly

The graphical user’s interfaces accessibility is a key condition for visual disabled peoples to use the computer applications. This paper presents the MOM (Meaningful Object Manager) software tool for accessibility analysis of graphical user interfaces. The accessibility depends not only on the tested applications, but also on assistive technologies. During this paper the functionalities of the MOM tools are described as well as the way these are used for accessibility evaluation. For each component are presented several case studies.

73-92

Authors: Ana Maria Marhan, Camelia Mihaela Popa

Recent studies focuses on topical issues related to Facebook, its use and impact on billions of young people´s social life who may accessing this website on a daily basis. User public profiles, as well as the content they distributed throughout the network can provide teachers and counsellors with a variety of socio-demographic information, along with information on skills, preferences, hobbies, and personality traits. Therefore, this paper critical examines the results of studies examining the online behaviour of users, particularly students, in order to analyse how certain personality traits may influence their preferences in using Facebook, as well as the relevance of such findings for the educational practice.