Discrete neighborhood representations and modified stacked generalization methods for distributed regression. Monge, R., Allende, H., Allende-Cid, H., & Moraga, C. R. Journal of Universal Computer Science, 21(6):842-855, 2015. abstract bibtex © J.UCSWhen distributed data sources have different contexts the problem of Distributed Regression becomes severe. It is the underlying law of probability that constitutes the context of a source. A new Distributed Regression System is presented, which makes use of a discrete representation of the probability density functions (pdfs). Neighborhoods of similar datasets are detected by comparing their approximated pdfs. This information supports an ensemble-based approach, and the improvement of a second level unit, as it is the case in stacked generalization. Two synthetic and six real data sets are used to compare the proposed method with other state-of-the-art models. The obtained results are positive for most datasets.
@article{84937821640,
abstract = "© J.UCSWhen distributed data sources have different contexts the problem of Distributed Regression becomes severe. It is the underlying law of probability that constitutes the context of a source. A new Distributed Regression System is presented, which makes use of a discrete representation of the probability density functions (pdfs). Neighborhoods of similar datasets are detected by comparing their approximated pdfs. This information supports an ensemble-based approach, and the improvement of a second level unit, as it is the case in stacked generalization. Two synthetic and six real data sets are used to compare the proposed method with other state-of-the-art models. The obtained results are positive for most datasets.",
number = "6",
year = "2015",
title = "Discrete neighborhood representations and modified stacked generalization methods for distributed regression",
volume = "21",
keywords = "Context-aware regression , Distributed machine learning , Similarity representation",
pages = "842-855",
journal = "Journal of Universal Computer Science",
author = "Monge, Raúl and Allende, Héctor and Allende-Cid, Héctor and Moraga, Claudio R."
}
Downloads: 0
{"_id":"c6BWunxtT4caGSHfC","bibbaseid":"monge-allende-allendecid-moraga-discreteneighborhoodrepresentationsandmodifiedstackedgeneralizationmethodsfordistributedregression-2015","downloads":0,"creationDate":"2017-04-03T15:16:25.244Z","title":"Discrete neighborhood representations and modified stacked generalization methods for distributed regression","author_short":["Monge, R.","Allende, H.","Allende-Cid, H.","Moraga, C. R."],"year":2015,"bibtype":"article","biburl":"https://1fichier.com/?yjj84y68k0","bibdata":{"bibtype":"article","type":"article","abstract":"© J.UCSWhen distributed data sources have different contexts the problem of Distributed Regression becomes severe. It is the underlying law of probability that constitutes the context of a source. A new Distributed Regression System is presented, which makes use of a discrete representation of the probability density functions (pdfs). Neighborhoods of similar datasets are detected by comparing their approximated pdfs. This information supports an ensemble-based approach, and the improvement of a second level unit, as it is the case in stacked generalization. Two synthetic and six real data sets are used to compare the proposed method with other state-of-the-art models. The obtained results are positive for most datasets.","number":"6","year":"2015","title":"Discrete neighborhood representations and modified stacked generalization methods for distributed regression","volume":"21","keywords":"Context-aware regression , Distributed machine learning , Similarity representation","pages":"842-855","journal":"Journal of Universal Computer Science","author":[{"propositions":[],"lastnames":["Monge"],"firstnames":["Raúl"],"suffixes":[]},{"propositions":[],"lastnames":["Allende"],"firstnames":["Héctor"],"suffixes":[]},{"propositions":[],"lastnames":["Allende-Cid"],"firstnames":["Héctor"],"suffixes":[]},{"propositions":[],"lastnames":["Moraga"],"firstnames":["Claudio","R."],"suffixes":[]}],"bibtex":"@article{84937821640,\n abstract = \"© J.UCSWhen distributed data sources have different contexts the problem of Distributed Regression becomes severe. It is the underlying law of probability that constitutes the context of a source. A new Distributed Regression System is presented, which makes use of a discrete representation of the probability density functions (pdfs). Neighborhoods of similar datasets are detected by comparing their approximated pdfs. This information supports an ensemble-based approach, and the improvement of a second level unit, as it is the case in stacked generalization. Two synthetic and six real data sets are used to compare the proposed method with other state-of-the-art models. The obtained results are positive for most datasets.\",\n number = \"6\",\n year = \"2015\",\n title = \"Discrete neighborhood representations and modified stacked generalization methods for distributed regression\",\n volume = \"21\",\n keywords = \"Context-aware regression , Distributed machine learning , Similarity representation\",\n pages = \"842-855\",\n journal = \"Journal of Universal Computer Science\",\n author = \"Monge, Raúl and Allende, Héctor and Allende-Cid, Héctor and Moraga, Claudio R.\"\n}\n\n","author_short":["Monge, R.","Allende, H.","Allende-Cid, H.","Moraga, C. R."],"key":"84937821640","id":"84937821640","bibbaseid":"monge-allende-allendecid-moraga-discreteneighborhoodrepresentationsandmodifiedstackedgeneralizationmethodsfordistributedregression-2015","role":"author","urls":{},"keyword":["Context-aware regression","Distributed machine learning","Similarity representation"],"downloads":0},"search_terms":["discrete","neighborhood","representations","modified","stacked","generalization","methods","distributed","regression","monge","allende","allende-cid","moraga"],"keywords":["context-aware regression","distributed machine learning","similarity representation"],"authorIDs":[],"dataSources":["mqyNxb9twi6rYAXFv"]}