Nuno Escudeiro

D-Confidence: an active learning strategy to reduce label disclosure complexity in the presence of imbalanced class distributions

In some classification tasks, such as those related to the automatic building and maintenance of text corpora, it is expensive to obtain labeled instances to train a classifier. In such circumstances it is common to have massive corpora where a few instances are labeled (typically a minority) while others are not. Semi-supervised learning techniques try to leverage the intrinsic information in unlabeled instances to improve classification models. However, these techniques assume that the labeled instances cover all the classes to learn which might not be the case. Moreover, when in the presence of an imbalanced class distribution, getting labeled instances from minority classes might be very costly, requiring extensive labeling, if queries are randomly selected. Active learning allows asking an oracle to label new instances, which are selected by criteria, aiming to reduce the labeling effort. D-Confidence is an active learning approach that is effective when in presence of imbalanced training sets. In this paper we evaluate the performance of d-Confidence in comparison to its baseline criteria over tabular and text datasets. We provide empirical evidence that d-Confidence reduces label disclosure complexity—which we have defined as the number of queries required to identify instances from all classes to learn—when in the presence of imbalanced data.

Enhancing students team work and communication skills in international settings

Our society is experiencing sudden changes in work organization in part due to the growing ease with which people can collaborate. Many successful cases of peer-to-peer models of organization arise and assume leading positions in world economy replacing, in many cases, the traditional hierarchical organization. People are evolving and interacting within heterogeneous teams composed by members from many different cultural groups and with distinct skills and backgrounds. Modern economy requires engineers to excel in collaborative and communication skills at an international setting. However, these competences are not usually addressed in most engineering curricula. We believe that in such a demanding and culturally diverse environment as the labour market is today, it is essential to promote team work and communication skills at an international and intercultural level. In the Multinational Undergraduate Team Work course, MUTW, students develop their capstone project as members of an international team while working at their home institutions. MUTW projects are to be developed by teams of final-year-undergraduate students from a multinational group of higher education institutions working to solve some engineering problem. Team members are geographically spread to assure heterogeneous teams and to promote international cooperation. This paradigm can be applied in any project/internship course unit. The results from the first edition are very encouraging supporting our initial hypothesis that MUTW significantly promotes students soft skills without requiring any change to prior degree curricula.

Efficient Coverage of Case Space with Active Learning

Collecting and annotating exemplary cases is a costly and critical task that is required in early stages of any classification process. Reducing labeling cost without degrading accuracy calls for a compromise solution which may be achieved with active learning. Common active learning approaches focus on accuracy and assume the availability of a pre-labeled set of exemplary cases covering all classes to learn. This assumption does not necessarily hold. In this paper we study the capabilities of a new active learning approach, d-Confidence, in rapidly covering the case space when compared to the traditional active learning confidence criterion, when the representativeness assumption is not met. Experimental results also show that d-Confidence reduces the number of queries required to achieve complete class coverage and tends to improve or maintain classification error.

Semi-automatic creation and maintenance of web resources with webtopic

In this paper we propose a methodology for automatically retrieving document collections from the web on specific topics and for organizing them and keeping them up-to-date over time, according to user specific persistent information needs. The documents collected are organized according to user specifications and are classified partly by the user and partly automatically. A presentation layer enables the exploration of large sets of documents and, simultaneously, monitors and records user interaction with these document collections. The quality of the system is permanently monitored; the system periodically measures and stores the values of its quality parameters. Using this quality log it is possible to maintain the quality of the resources by triggering procedures aimed at correcting or preventing quality degradation.

Slovenian students on projects/internships

PRAXIS is the European Network aiming to setup the European Center for Project/Internship (IP) Excellence, which is developed for needs of project oriented education and between others also support internationalization at home and mobility of students abroad. Project/Internship course units are particularly relevant to engineering degrees. It is through this type of courses that students have the chance to practice their technical skills in a real-world-like setting and experience soft skills that are a key factor for employability. Besides students and teachers also companies and HEI (Higher Education Institution) are included and integrated into PRAXIS and new connections are possible. For needs of PRAXIS project, we did a PI survey in which students, HEI and companies where involved. We decided to have analysis for Siovenian student separately on the base number of completed questionnaires/partners ratio, which was the highest between all 31 participating countries.

Virtual Sign – A Real Time Bidirectional Translator of Portuguese Sign Language

Abstract Promoting equity, equal opportunities to all and social inclusion of people with disabilities is a concern of modern societies at large and a key topic in the agenda of European Higher Education. Despite all the progress, we cannot ignore the fact that the conditions provided by the society for the deaf are still far from being perfect. The communication with deaf by means of written text is not as efficient as it might seem at first. In fact, there is a very deep gap between sign language and spoken/written language. The vocabulary, the sentence construction and the grammatical rules are quite different among these two worlds. These facts bring significant difficulties in reading and understanding the meaning of text for deaf people and, on the other hand, make it quite difficult for people with no hearing disabilities to understand sign language. The deployment of tools to assist the daily communication, in schools, in public services, in museums and other, between deaf people and the rest may be a significant contribution to the social inclusion of the deaf community. The work described in this paper addresses the development of a bidirectional translator between Portuguese Sign Language and Portuguese text. The translator from sign language to text resorts to two devices, namely the Microsoft Kinect and 5DT Sensor Gloves in order to gather data about the motion and shape of the hands. The hands configurations are classified using Support Vector Machines. The classification of the movement and orientation of the hands are achieved through the use of Dynamic Time Warping algorithm. The translator exhibits a precision higher than 90%. In the other direction, the translation of Portuguese text to Portuguese Sign Language is supported by a 3D avatar which interprets the entered text and performs the corresponding animations.

Evaluating educational games in mobile platforms

In this paper we present a quality model based on the software engineering paradigms, which is applied to evaluate digital learning content, specifically, games for mobile platforms. The model is called Quantitative Evaluation Framework (QEF), and measures system quality throughout its development life cycle. QEF evaluates the educational software quality, particularly the digital learning content (ISO 9126 is the standard of reference) in a three-dimensional space. Every dimension aggregates a set of factors. A factor is a component that represents the system performance from a particular point of view. The quality of a given system is defined and evaluated in our tri-dimensional Cartesian quality space in relation to a hypothetical ideal system represented in our quality space by the coordinates (1, 1, 1).

Developing Undergraduate Projects in Multinational Teams to Enhance Employability

Our society is experiencing sudden changes in work organization in part due to the growing ease with which people can collaborate. Many successful cases of peer-to-peer models of organization arise and assume leading positions in world economy replacing, in many cases, the traditional hierarchical organization. People are evolving and interacting within heterogeneous teams composed by members from many different cultural groups and with distinct skills and backgrounds. Modern economy requires engineers to excel in collaborative and communication skills at an international setting. However, these competences are not usually addressed in most engineering curricula. We believe that in such a demanding and culturally diverse environment as the labour market is today, it is essential to promote team work and communication skills at an international and intercultural level. In the Multinational Undergraduate Team Work course, MUTW, students develop their capstone project as members of an international team while working at their home institutions. MUTW projects are to be developed by teams of final-year-undergraduate students from a multinational group of higher education institutions working to solve some engineering problem. Team members are geographically spread to assure heterogeneous teams and to promote international cooperation. This paradigm can be applied in any project/internship course unit. The results from the first edition are very encouraging supporting our initial hypothesis that MUTW significantly promotes students soft skills without requiring any change to prior degree curricula.

Digital Assisted Communication.

The communication with the deaf community can prove to be very challenging without the use of sign language. There is a considerable difference between sign and written language as they differ in both syntax and semantics. The work described in this paper addresses the development of a bidirectional translator between several sign languages and their respective text, as well as the evaluation methods and results of those tools. A multiplayer game is using the translator is also described on this paper. The translator from sign language to text employs two devices, namely the Microsoft Kinect and 5DT Sensor Gloves in order to gather data about the motion and shape of the hands. This translator is being adapted to allow the communication with the blind as well. The Quantitative Evaluation Framework (QEF) and the ten-fold cross-validation methods were used to evaluate the project and show promising results. Also, the product goes through a validation process by sign language experts and deaf users who provide their feedback answering a questionnaire. The translator exhibits a precision higher than 90% and the projects overall quality rates are close to 90% based on the QEF.

Recognition of hand configuration: A critical factor in automatic sign language translation

Identifying hand configuration is a critical feature of sign language translation. In this paper, we describe our approach to recognize hand configurations in real time with the purpose of providing accurate predictions to be used in automatic sign language translation. To capture the hand configuration we rely on data gloves with 14 sensors that measure finger joints bending. These inputs are sampled at a frequency of 100Hz and fed to a classifier that predicts the current hand configuration. The classification model is created from an annotated sample of hand configurations previously acquired. We expect this approach to be accurate and robust in the sense that the performance of the classification model should not vary significantly when the classifier is being used by one or another user. The results from our experimental evaluation show that there is a very high accuracy, meaning that data gloves are a good approach to capture the descriptive features of hand configurations. However, the robustness of such an approach is not as good as desirable since the accuracy of the classifier depends on the user, i.e., the accuracy is high when the classifier is used by a user who trained it but decreases in other cases.