Mostra el registre d'ítem simple
On the predictive power of meta-features in OpenML
dc.contributor.author | Bilalli, Besim |
dc.contributor.author | Abelló Gamazo, Alberto |
dc.contributor.author | Aluja Banet, Tomàs |
dc.contributor.other | Universitat Politècnica de Catalunya. Departament d'Enginyeria de Serveis i Sistemes d'Informació |
dc.contributor.other | Universitat Politècnica de Catalunya. Departament d'Estadística i Investigació Operativa |
dc.date.accessioned | 2018-01-26T08:33:16Z |
dc.date.available | 2018-01-26T08:33:16Z |
dc.date.issued | 2017-12-20 |
dc.identifier.citation | Bilalli, B., Abello, A., Aluja, T. On the predictive power of meta-features in OpenML. "International journal of applied mathematics and computer science", 20 Desembre 2017, vol. 27, núm. 4, p. 697-712. |
dc.identifier.issn | 1641-876X |
dc.identifier.uri | http://hdl.handle.net/2117/113229 |
dc.description.abstract | The demand for performing data analysis is steadily rising. As a consequence, people of different profiles (i.e., non-experienced users) have started to analyze their data. However, this is challenging for them. A key step that poses difficulties and determines the success of the analysis is data mining (model/algorithm selection problem). Meta-learning is a technique used for assisting non-expert users in this step. The effectiveness of meta-learning is, however, largely dependent on the description/characterization of datasets (i.e., meta-features used for meta-learning). There is a need for improving the effectiveness of meta-learning by identifying and designing more predictive meta-features. In this work, we use a method from exploratory factor analysis to study the predictive power of different meta-features collected in OpenML, which is a collaborative machine learning platform that is designed to store and organize meta-data about datasets, data mining algorithms, models and their evaluations. We first use the method to extract latent features, which are abstract concepts that group together meta-features with common characteristics. Then, we study and visualize the relationship of the latent features with three different performance measures of four classification algorithms on hundreds of datasets available in OpenML, and we select the latent features with the highest predictive power. Finally, we use the selected latent features to perform meta-learning and we show that our method improves the meta-learning process. Furthermore, we design an easy to use application for retrieving different meta-data from OpenML as the biggest source of data in this domain. |
dc.format.extent | 16 p. |
dc.language.iso | eng |
dc.rights | Attribution-NonCommercial-NoDerivs 3.0 Spain |
dc.rights.uri | http://creativecommons.org/licenses/by-nc-nd/3.0/es/ |
dc.subject | Àrees temàtiques de la UPC::Informàtica::Aplicacions de la informàtica::Aplicacions informàtiques a la física i l‘enginyeria |
dc.subject.lcsh | OpenMP (Application program interface) |
dc.subject.other | feature extraction |
dc.subject.other | feature selection |
dc.subject.other | meta-learning |
dc.title | On the predictive power of meta-features in OpenML |
dc.type | Article |
dc.subject.lemac | Interfícies de programació d'aplicacions (Programari) |
dc.contributor.group | Universitat Politècnica de Catalunya. inSSIDE - integrated Software, Service, Information and Data Engineering |
dc.contributor.group | Universitat Politècnica de Catalunya. LIAM - Laboratori de Modelització i Anàlisi de la Informació |
dc.identifier.doi | 10.1515/amcs-2017-0048 |
dc.description.peerreviewed | Peer Reviewed |
dc.relation.publisherversion | https://www.degruyter.com/view/j/amcs.2017.27.issue-4/amcs-2017-0048/amcs-2017-0048.xml |
dc.rights.access | Open Access |
local.identifier.drac | 21872765 |
dc.description.version | Postprint (published version) |
local.citation.author | Bilalli, B.; Abello, A.; Aluja, T. |
local.citation.publicationName | International journal of applied mathematics and computer science |
local.citation.volume | 27 |
local.citation.number | 4 |
local.citation.startingPage | 697 |
local.citation.endingPage | 712 |
Fitxers d'aquest items
Aquest ítem apareix a les col·leccions següents
-
Articles de revista [113]
-
Articles de revista [50]
-
Articles de revista [222]
-
Articles de revista [56]
-
Articles de revista [719]