Osvaldo Pacheco

Classifier for drinking water quality in real time

Real time features are critical for automatic assessment of Drinking Water Quality (DWQ). This paper explores the use of real time features to feed machine learning classifiers for DWQ. Two different representative datasets were used from: a) The Provincial Water Quality Monitoring Network from Ontario, Canada and b) National Hydrologic Information System from Central Region of Portugal. The procedure followed in this study was: (1) automatically computing a Water Quality Index to classify the datasets elements in five classes (excellent, good, medium, bad and very bad) using the Kumar method; (2) selecting best performed real time features on results of classified datasets; and (3) exploring machine learning algorithms (e.g. Decision Trees, Artificial Neural Networks and k-Nearest Neighbor) for producing DWQ classifiers. In this work, we perform the classification of two classes (good and medium) out of the five possible categories, due to the absence of vectors in the datasets.

User Identification: A Key Factor for Elderly Viewers to Benefit from Interactive Television Services

The offer of interactive television services is becoming more and more frequent in regular households allowing a more customizable way of watching TV. However, in a multi-viewer environment a truly personalized TV experience could not merely rely on the identification of the STB that users have at their homes. For this purpose, a non-intrusive user identification technique is especially relevant, namely when senior viewers are at stake. In this work we are studying different approaches to perform an adequate identification of senior viewers so they can benefit from a series of interactive services targeted to their needs. To achieve this aim, a set of exploratory interviews have been carried out, and based on its results, a prototype with several identification technologies has been developed. With this prototype and through a set of usability tests we were able to collect relevant opinions about the type of added value TV services that this target audience desires. In addition, it was also possible to understand points of view that lead us to conclude that the user context involves a complex trade-off that influences the choice of the most suitable identification system.

Forecasting British Tourist Inflows to Portugal Using Google Trends Data

Purpose—The purpose of this paper is to explore the Google Trends (GT) data in order to understand the behavior and interests of British tourists in Portugal as a tourist destination and to verify if the GT data correlates with the tourism official data of Portugal. Furthermore, it will investigate if GT data can improve forecasts on the arrival of British tourists to Portugal. Design/methodology/approach—We used GT data on a set of search terms to predict the demand for hotel establishments by UK residents in Portugal and employed the Autoregressive Integrated Moving Average (ARIMA) model and Transfer Function (TF) to evaluate the usefulness of this data. Furthermore, we correlated the GT data with official tourism data of Portugal. Findings—The TF models outperformed their ARIMA counterparts, meaning that the TF models which considered the GT index produced more accurate forecasts. Practical implications—The paper contributes to increase the knowledge on the potential of Google-based search data in order to understand the behaviour patterns of predicted British travelers to Portugal and help to predict the British tourist inflows to Portugal. Originality/value—The paper is novel because it is the first in the field of hospitality and tourism to predict British tourists inflows to Portugal and it is a unique paper in this area that used several keywords in order to define a tourist destination.

Effects of curriculum adjustments on first-year programming courses: students performance and achievment

1 Isabel Huet e Silva, Science Education Department, University of Aveiro, iza@dce.ua.pt 2 Jose Tavares, Science Education Department, University of Aveiro, tavares@dce.ua.pt 3 Osvaldo Pacheco, Electronic and Telecommunication Department, University of Aveiro, osvaldo@det.ua.pt Abstract Researchers of the Science Education Department have been working together for the past two years with faculty from the Electronic and Telecommunication Department aiming to explain students’ failure and drop out rate in a first-year undergraduate programming course at the University of Aveiro, Portugal. Two main variables were significantly correlated with the students’ failure: the curriculum organization of the course and the teaching methods. In this paper we are concerned with analyzing the curriculum redesign in the area of Informatics, namely at a introductory programming course that took place in 2001/02 and whether it has influenced students’ achievement and motivation for the course.

Classification of water for production using parameters in real time

In this paper, a new classification method for production water is proposed, based on so real-time measured parameters. The classification method consists of three steps: 1) An initial classification of the Water Quality Index is computed using the method proposed by KUMAR; 2) Feature selection based on random forest (specifically based on the method varSelRF); and 3) Training of classifiers using different configurations of heuristic decision trees. A total of 4 datasets (5090 instances of 8 features each) representative of water samples from Portugal, Canada, Mexico, and Romania were used for method validation. The dataset was group in two families of different classes: binary (good and regular water) and multiclass (good, regular and bad water). Final classification accuracy reached 94.85% for the binary family and 91.73% for the multiclass family. The contribution consists of a continuous monitoring system to detect (in real time) dramatic changes in water quality and provide tools for historical studies behaviour in strategic points.