Revista Romana de Interactiune Om-Calculator

307-322

Authors: Alexandra Negoescu, Cristian Mihăescu, Ilya Shmorgun, Mihai Mocanu

Depending on the size and functionality, kiosks have become an outstanding way for improving customer service and gaining users’ loyalty. In the educational domain most of the available kiosks contain just general knowledge about the university in question or information about campus and local buildings. In addition, the majority of kiosks available in universities don't have support for touch or multi-touch, most probably due to the high prices involved. We proposed to design and build our own version of an interactive touch- kiosk using up-to-date technologies that best fit the current needs of TallinnUniversity. Our aim was to design and implement an interface that should be completely dedicated and self-contained, thus minimizing user awareness of operating system components or software and hardware infrastructure. Moreover, we proposed the usage of technologies such as Windows 8 operating system and infrared-based touch-screens. Touch the Exchange is a digital environment running on a touch-screen kiosk that provides essential support for persons interested in international exchange possibilities. The aim of the project is to encourage local students and university staff members to take advantage of the opportunities available to them. This paper presents a description of the hardware and software technologies used within the project, as well as the reason why these specific implementation decisions have been chosen. Also, special attention is also given to the design process of the interface, its evaluation and testing and to the manner in which human-computer communication is facilitated. The whole process: the initial design concept, evaluation of the needed components, modelling, certification and construction of the prototype as well as its production are outlined in this article. The result, an attractive, modern and user-friendly system that ensures direct access to information can serve as a model for the educational institutions interesed in developing similar devices with minimum costs.

339-356

Authors: Lucia Larise Stavarache, Mihai Dascalu, Stefan Trausan-Matu, Philippe Dessus

One of the most challenging tasks in natural language processing is related to the evaluation of textual complexity and of cohesion. Thus, automatic analyzers represent a special interest for their ability of predicting textual complexity having as basis: syntactic, sematic, morphological and lexical factors. Nevertheless, the elements used for evaluating textual complexity are tightly connected to scope and context. For this reason a purely technical analysis based only on individual metrics, without taking into account the constraints of psychology, human model, age and motivation is insufficient in order to make a proper prediction. Additionally, there are third party aspects which are correlated with the analysis of complexity like: stages of acquisition, fluency or accuracy. To resume, textual complexity metrics represent important indicators which help to better understand online texts, publications, books, papers or other forms of written textual representations.