Rank Aggregation for Non-stationary Data Streams

Ekhine Irurozki, Aritz Perez, Jesus Lobo, Javier Del Ser

Producción científica: Capítulo del libro/informe/acta de congresoContribución a la conferenciarevisión exhaustiva

1 Descargas (Pure)

Resumen

The problem of learning over non-stationary ranking streams arises naturally, particularly in recommender systems. The rankings represent the preferences of a population, and the non-stationarity means that the distribution of preferences changes over time. We propose an algorithm that learns the current distribution of ranking in an online manner. The bottleneck of this process is a rank aggregation problem. We propose a generalization of the Borda algorithm for non-stationary ranking streams. As a main result, we bound the minimum number of samples required to output the ground truth with high probability. Besides, we show how the optimal parameters are set. Then, we generalize the whole family of weighted voting rules (the family to which Borda belongs) to situations in which some rankings are more reliable than others. We show that, under mild assumptions, this generalization can solve the problem of rank aggregation over non-stationary data streams.
Idioma originalInglés
Título de la publicación alojadaunknown
EditoresNuria Oliver, Fernando Pérez-Cruz, Stefan Kramer, Jesse Read, Jose A. Lozano
EditorialSpringer
Páginas297-313
Número de páginas17
Volumen12977
ISBN (versión impresa)9783030865221
DOI
EstadoPublicada - 11 sept 2021
EventoEuropean Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML PKDD 2021 - Virtual, Online
Duración: 13 sept 202117 sept 2021

Serie de la publicación

Nombre0302-9743

Conferencia

ConferenciaEuropean Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML PKDD 2021
CiudadVirtual, Online
Período13/09/2117/09/21

Palabras clave

  • Preference learning
  • Rank aggregation
  • Borda
  • Evolving preferences
  • Voting
  • Concept drift

Project and Funding Information

  • Funding Info
  • This work is partially funded by the Industrial Chair “Data science & Artificial Intelligence for Digitalized Industry & Services” from Telecom Paris (France), _x000D_ the Basque Government through the BERC 2018–2021 and the Elkartek program (KK-2018/00096, KK-2020/00049), _x000D_ and by the Spanish Government excellence accreditation Severo Ochoa SEV-2013-0323 (MICIU) and the project TIN2017-82626-R (MINECO). _x000D_ J. Del Ser also acknowledges funding support from the Basque Government (Consolidated Research Gr. MATHMODE, IT1294-19).

Huella

Profundice en los temas de investigación de 'Rank Aggregation for Non-stationary Data Streams'. En conjunto forman una huella única.

Citar esto