An Information Retrieval Approach for Finding Dependent Subspaces of Multiple Views

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussaConference contributionScientificvertaisarvioitu

1 Sitaatiot (Scopus)

Abstrakti

Finding relationships between multiple views of data is essential both in exploratory analysis and as pre-processing for predictive tasks. A prominent approach is to apply variants of Canonical Correlation Analysis (CCA), a classical method seeking correlated components between views. The basic CCA is restricted to maximizing a simple dependency criterion, correlation, measured directly between data coordinates.We introduce a new method that finds dependent subspaces of views directly optimized for the data analysis task of neighbor retrieval between multiple views. We optimize mappings for each view such as linear transformations to maximize cross-view similarity between neighborhoods of data samples. The criterion arises directly from the well-defined retrieval task, detects nonlinear and local similarities, measures dependency of data relationships rather than only individual data coordinates, and is related to well understood measures of information retrieval quality. In experiments the proposed method outperforms alternatives in preserving cross-view neighborhood similarities, and yields insights into local dependencies between multiple views.
AlkuperäiskieliEnglanti
OtsikkoMachine Learning and Data Mining in Pattern Recognition
Alaotsikko13th International Conference, MLDM 2017 New York, NY, USA, July 15 – 20, 2017, Proceedings
ToimittajatPetra Perner
Sivut1-16
Sivumäärä15
ISBN (elektroninen)978-3-319-62416-7
DOI - pysyväislinkit
TilaJulkaistu - heinäkuuta 2017
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisuussa
TapahtumaInternational Conference on Machine Learning and Data Mining - New York, Yhdysvallat
Kesto: 15 heinäkuuta 201720 heinäkuuta 2017
Konferenssinumero: 13

Julkaisusarja

NimiLecture Notes in Computer Science
KustantajaSpringer
Numero10358
ISSN (painettu)0302-9743
ISSN (elektroninen)1611-3349

Conference

ConferenceInternational Conference on Machine Learning and Data Mining
LyhennettäMLDM
MaaYhdysvallat
KaupunkiNew York
Ajanjakso15/07/201720/07/2017

Sormenjälki

Sukella tutkimusaiheisiin 'An Information Retrieval Approach for Finding Dependent Subspaces of Multiple Views'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

Siteeraa tätä