Abstract
Several predictive systems are nowadays vital for operations and decision support. The quality of these systems is most of the time defined by their average accuracy which has low or no information at all about the estimated error of each individual prediction. In these cases, users should be allowed to associate a measure of reliability to each prediction. However, with the advent of data streams, batch state-of-the-art reliability estimates need to be redefined. In this chapter we adapt and evaluate five empirical measures for online reliability estimation of individual predictions: similarity-based (k-NN) error, local sensitivity (bias and variance) and online bagging predictions (bias and variance). Evaluation is performed with a neural network base model on two different problems, with results showing that online bagging and k-NN estimates are consistently correlated with the error of the base model. Furthermore, we propose an approach for correcting individual predictions based on the CNK reliability estimate. Evaluation is done on a real-world problem (prediction of the electricity load for a selected European geographical region), using two different regression models: neural network and the k nearest neighbors algorithm. Comparison is performed with corrections based on the Kalman filter. The results show that our method performs better than the Kalman filter, significantly improving the original predictions to more accurate values.

Abstract

Abstract
Entrepreneurship is widely recognized as one of the basic skills to be acquired through a life-long learning. The European Union, under the guidance of the Oslo Agenda, promotes several initiatives to develop entrepreneurship culture in Europe. Education can make a significant contribution to entrepreneurship, encouraging the development of entrepreneurial attitudes and skills in young people. Serious Games are presently recognised as having an important role and potential in education and social networks emerged in the last years as the platform preferred by many, especially young people, to socialize, play games and even learn. This paper presents the PLAYER project, in which a game was developed and implemented as a Facebook application, to enable learning entrepreneurial skills progressively, by guiding users to develop a business idea in the form of a business plan.

Abstract
Lately, with the goal of improving road safety, car makers developed and commercialised some Advanced Driver Assistance Systems (ADAS) which, through the detection of blind spot areas on the vehicle's sides, could help the drivers during the overtaking and the change lane task. Despite the possible benefits to reduce lateral crashes, the overall impact on road safety of such systems have not been deeply studied yet; notably, despite some researches have been carried out, there is a lack of studies regarding the long-term usage and drivers' acceptance of those systems. In order to fill the research gap, a methodology, based on the combination of focus groups interviews, questionnaires and a small-scale field operational test (FOT), has been designed in this study; such a methodology aims at evaluating drivers' acceptance of Blind Spot Information System (R) and at proposing some ideas to improve the usability and user-friendliness of this (or similar) device in their future development.

Abstract

Abstract
In recent years, mass adoption of increasingly powerful mobile devices and ubiquitous communication networks have paved the way to smart environments. Such environments allow for the collection of user and environment data with the final goal to improve users' experience. In this context a number of opportunities and challenges are presented to Human Computer Interaction. This poster explores Quality of Experience, a subjective aspect of interaction, informally defined as the degree to which a system meets users' expectations. Furthermore, a mobile application was developed for the collection of user and environment data and delivery of personalised services in the context of Public Transport. This application will be used in a real-world environment, to further investigate the factors that have an influence on User eXperience, as well as the delivery of relevant services with the potential to enhance users' journeys while in transit.