Matches in SemOpenAlex for { <https://semopenalex.org/work/W2585832308> ?p ?o ?g. }
- W2585832308 abstract "Forecasts combinations normally use point forecasts that were obtained from different models or sources ([1], [2],[3]). This paper explores the incorporation of internal structure parameters of feed-forward neural network (NN) models as anapproach to combine their forecasts via ensembles. First, the generated NN models that could be part of the ensembles are subjectto a clustering algorithm that uses the structure parameters and, from each of the clusters obtained, a small set of models is selectedand their forecasts are combined in a two-stage procedure. Secondly, in an alternative and simpler implementation, a subset of the generated NN models is selected by using several reference points in the model structure parameter space. The choice of thereference points is optimised through a genetic algorithm and the models selected are averaged. Hourly electricity demand timeseries is used to assess multi-step ahead forecasting performance for up to a 12 hours horizon. Results are compared against severalstatistical benchmarks, the average of the individual forecasts and the best models in the ensembles. Results show that the cluster-based (CB) structural combinations do better than the genetic algorithm (GA) structural combinations in outperforming the average forecast, which is the traditional point forecast from an ensemble." @default.
- W2585832308 created "2017-02-10" @default.
- W2585832308 creator A5047727633 @default.
- W2585832308 creator A5065033974 @default.
- W2585832308 date "2016-12-01" @default.
- W2585832308 modified "2023-09-24" @default.
- W2585832308 title "Structural Combination of Neural Network Models" @default.
- W2585832308 cites W1541843935 @default.
- W2585832308 cites W1605276600 @default.
- W2585832308 cites W1964298122 @default.
- W2585832308 cites W1972860770 @default.
- W2585832308 cites W1986478348 @default.
- W2585832308 cites W2005256561 @default.
- W2585832308 cites W2007272376 @default.
- W2585832308 cites W2008869685 @default.
- W2585832308 cites W2014349106 @default.
- W2585832308 cites W2037671496 @default.
- W2585832308 cites W2037677918 @default.
- W2585832308 cites W2061082730 @default.
- W2585832308 cites W2072443317 @default.
- W2585832308 cites W2075795701 @default.
- W2585832308 cites W2079537749 @default.
- W2585832308 cites W2082332823 @default.
- W2585832308 cites W2092436092 @default.
- W2585832308 cites W2100128988 @default.
- W2585832308 cites W2101265116 @default.
- W2585832308 cites W2102201073 @default.
- W2585832308 cites W2103414828 @default.
- W2585832308 cites W2104165437 @default.
- W2585832308 cites W2106390255 @default.
- W2585832308 cites W2108537597 @default.
- W2585832308 cites W2115055351 @default.
- W2585832308 cites W2122258997 @default.
- W2585832308 cites W2126186680 @default.
- W2585832308 cites W2127404166 @default.
- W2585832308 cites W2131534673 @default.
- W2585832308 cites W2135293965 @default.
- W2585832308 cites W2135995262 @default.
- W2585832308 cites W2145732764 @default.
- W2585832308 cites W2145833756 @default.
- W2585832308 cites W2146097858 @default.
- W2585832308 cites W2149491889 @default.
- W2585832308 cites W2149905014 @default.
- W2585832308 cites W2569241116 @default.
- W2585832308 cites W27682692 @default.
- W2585832308 cites W1993387039 @default.
- W2585832308 doi "https://doi.org/10.1109/icdmw.2016.0064" @default.
- W2585832308 hasPublicationYear "2016" @default.
- W2585832308 type Work @default.
- W2585832308 sameAs 2585832308 @default.
- W2585832308 citedByCount "1" @default.
- W2585832308 countsByYear W25858323082019 @default.
- W2585832308 crossrefType "proceedings-article" @default.
- W2585832308 hasAuthorship W2585832308A5047727633 @default.
- W2585832308 hasAuthorship W2585832308A5065033974 @default.
- W2585832308 hasBestOaLocation W25858323082 @default.
- W2585832308 hasConcept C11413529 @default.
- W2585832308 hasConcept C119857082 @default.
- W2585832308 hasConcept C124101348 @default.
- W2585832308 hasConcept C154945302 @default.
- W2585832308 hasConcept C177264268 @default.
- W2585832308 hasConcept C199360897 @default.
- W2585832308 hasConcept C2524010 @default.
- W2585832308 hasConcept C28719098 @default.
- W2585832308 hasConcept C33923547 @default.
- W2585832308 hasConcept C41008148 @default.
- W2585832308 hasConcept C50644808 @default.
- W2585832308 hasConcept C73555534 @default.
- W2585832308 hasConcept C8880873 @default.
- W2585832308 hasConceptScore W2585832308C11413529 @default.
- W2585832308 hasConceptScore W2585832308C119857082 @default.
- W2585832308 hasConceptScore W2585832308C124101348 @default.
- W2585832308 hasConceptScore W2585832308C154945302 @default.
- W2585832308 hasConceptScore W2585832308C177264268 @default.
- W2585832308 hasConceptScore W2585832308C199360897 @default.
- W2585832308 hasConceptScore W2585832308C2524010 @default.
- W2585832308 hasConceptScore W2585832308C28719098 @default.
- W2585832308 hasConceptScore W2585832308C33923547 @default.
- W2585832308 hasConceptScore W2585832308C41008148 @default.
- W2585832308 hasConceptScore W2585832308C50644808 @default.
- W2585832308 hasConceptScore W2585832308C73555534 @default.
- W2585832308 hasConceptScore W2585832308C8880873 @default.
- W2585832308 hasLocation W25858323081 @default.
- W2585832308 hasLocation W25858323082 @default.
- W2585832308 hasOpenAccess W2585832308 @default.
- W2585832308 hasPrimaryLocation W25858323081 @default.
- W2585832308 hasRelatedWork W1537908023 @default.
- W2585832308 hasRelatedWork W1701194797 @default.
- W2585832308 hasRelatedWork W1743046522 @default.
- W2585832308 hasRelatedWork W1988437555 @default.
- W2585832308 hasRelatedWork W2016166233 @default.
- W2585832308 hasRelatedWork W2041811496 @default.
- W2585832308 hasRelatedWork W2080819414 @default.
- W2585832308 hasRelatedWork W2122561198 @default.
- W2585832308 hasRelatedWork W2143322324 @default.
- W2585832308 hasRelatedWork W2171715293 @default.
- W2585832308 hasRelatedWork W2327285060 @default.
- W2585832308 hasRelatedWork W2484759140 @default.
- W2585832308 hasRelatedWork W2585510649 @default.
- W2585832308 hasRelatedWork W2788949274 @default.