LIAAD - Book Chapters

Permanent URI for this collection

http://localhost:4000/handle/123456789/198

Browse

Browsing LIAAD - Book Chapters by Author "Carlos Ferreira"
  • Item
    Bus Bunching Detection by Mining Sequences of Headway Deviations
    ( 2012) Jorge Freire ; Luís Moreira Matias ; Carlos Ferreira ; João Gama ; João Mendes Moreira
    In highly populated urban zones, it is common to notice headway deviations (HD) between pairs of buses. When these events occur in a bus stop, they often cause bus bunching (BB) in the following bus stops. Several proposals have been suggested to mitigate this problem. In this paper, we propose to find BBS (Bunching Black Spots) - sequences of bus stops where systematic HD events cause the formation of BB. We run a sequence mining algorithm, named PrefixSpan, to find interesting events available in time series. We prove that we can accurately model the BB trip usual pattern like a frequent sequence mining problem. The subsequences proved to be a promising way of identify the route' schedule points to adjust in order to mitigate such events.
  • Item
    Identifying Relationships in Transactional Data
    ( 2012) João Gama ; Melissa Rodrigues ; Carlos Ferreira
    Association rules is the traditional way used to study market basket or transactional data. One drawback of this analysis is the huge number of rules generated. As a complement to Association Rules, Association Rules Network (ARN), based on Social Network Analysis (SNA) has been proposed by several researchers. In this work we study a real market basket analysis problem, available in a Belgian supermarket, using ARNs. We learn ARNs by considering the relationships between items that appear more often in the consequent of the Association Rules. Moreover, we propose a more compact variant of ARNs: the Maximal Itemsets Social Network (MISN). In order to assess the quality of these structures, we compute SNA based metrics, like weighted degree and utility of community.
  • Item
    Probabilistic ramp detection and forecasting for wind power prediction
    ( 2012) Carlos Ferreira ; Audun Botterud ; João Gama ; Vladimiro Miranda
    This paper presents a new approach to the critical problem of detecting or forecasting ramping events in the context of wind power prediction. The novelty of the model relies on departing from the probability density function estimated for the wind power and building a probabilistic representation of encountering, at each time step, a ramp event according to some definition. The model allows the assignment of a probability value to each possible magnitude of a predicted ramp and its worth is assessed by several metrics including ROC curves.