http://hdl.handle.net/10256/19666 Tue, 13 Jan 2026 06:52:59 GMT 2026-01-13T06:52:59Z http://hdl.handle.net/10256/24738 Frequent patterns of childhood overweight from longitudinal data on parental and early-life of infants health López Ibáñez, Beatriz; Galera, David; López-Bermejo, Abel; Bassols Casadevall, Judit Childhood obesity is considered one of the main public health concerns. Research in the field of obesity detection and prevention is moving towards promising solutions thanks to the use of Artificial Intelligence applied to data from cohorts of children. Previous studies have analyzed the data without taking into account the relationship of data regarding when they are collected. In this work, frequent pattern mining is used to find the risk factors of childhood obesity, taking into account the relationship among the data gathered in different visits. The experiments carried out on the data collected from 386 children from Girona and Figueres (Spain) demonstrate the relevance of discriminant frequent patterns for childhood overweight prediction Article relacionat amb la comunicació que es presentarà a AIME 2024. 22nd International Conference on Artificial Intelligence in Medicine: Salt Lake City, USA: July 9-12 Mon, 01 Jan 2024 00:00:00 GMT http://hdl.handle.net/10256/24738 2024-01-01T00:00:00Z http://hdl.handle.net/10256/23165 Forecasting Return Visits of the Emergency Department Mordvanyuka, Natalia; Torrent-Fontbona, Ferran; Inoriza, José María; López Ibáñez, Beatriz Emergency Department (ED) revisits are aggravating ED patient overcrowding. |- ADAPT The use of arti cial intelligence techniques to nd out which Single Nucleotide Polymorphisms (SNPs) promote the development of a disease is one of the features of medical research, as such techniques may potentially aid early diagnosis and help in the prescription of preventive measures. In particular, the aim is to help physicians to identify the relevant SNPs related to Type 2 diabetes, and to build a decision-support tool for risk prediction. Methods: In order to support ED manager, we propose the NearMiss algorithm in combination with an ANN approach to forecast ED revisits. { ADAPT We use the Random Forest (RF) technique in order to search for the most important attributes (SNPs) related to diabetes, giving a weight (degree of importance), ranging between 0 and 1, to each attribute. Support vector machines and logistic regression have also been used since they are two other machine learning techniques that are well-established in the health community. Their performance has been compared to that achieved by RF. Furthermore, the relevance of the attributes obtained through the use of RF has then been used to perform predictions with k nearest neighbour method weighting attributes in the similarity measure according to the relevance of the attributes with RF. Results: The method is applied to a dataset of 12 years and outperforms previous works in the literature with a prediction AUC of 89.6%. { ADAPT Testing is performed on a set of 677 subjects. RF is able to handle the complexity of features' interactions, over tting, and unknown attribute values, providing the SNPs' relevance with an up to 0.89 area under the ROC curve in terms of risk prediction. RF outperforms all the other tested machine learning techniques in terms of prediction accuracy, and in terms of the stability of the estimated relevance of the attributes. Conclusions: TODO - ADAPT The random forest is a useful method for learning predictive models and the relevance of SNPs without any underlying assumption Sat, 20 Jul 2019 00:00:00 GMT http://hdl.handle.net/10256/23165 2019-07-20T00:00:00Z http://hdl.handle.net/10256/22958 Energy-aware multi-mode resource constrained project scheduling under time dependent electricity costs and user compromised consumption Torrent-Fontbona, Ferran; López Ibáñez, Beatriz Future smart grid involves the monitoring and control of the energy consumption pro le of each consumer with demand-side strategies aiming to incentivise changes in consumers' pro les like time-dependent energy prices and compromised load patterns. However, demand-side management strategies need consumers capable to respond to the incentives. As a consequence, the project scheduling problem, which consists of a set of activities that has to be scheduled subject to precedence and resource constraints, need to be reviewed to consider the new challenges posed by smart grids. In this paper we model the multi-mode project scheduling problem under time-dependent energy prices and compromised load patterns to support decision making with energy-aware issues. We carried out experimentation based on real-based simulated scenarios to show the consequences of the proposed model Thu, 05 Nov 2015 00:00:00 GMT http://hdl.handle.net/10256/22958 2015-11-05T00:00:00Z