Multivariate Correlation Analysis for Supervised Feature Selection in High-Dimensional Data

Thumbnail
Author
Shekar, Arvind Kumar
Type of Scholarly Publication
Dissertation
Date of Exam
11.12.2019
Date of Publication
12.03.2020
Advisor
Müller, Emmanuel
Co-Referee
Hüllermeier, Eyke
Involved Institutions
Rheinische Friedrich-Wilhelms-Universität Bonn
Metadata
Show full item record
Citable Links
Abstract
The main theme of this dissertation focuses on multivariate correlation analysis on different data types and we identify and define various research gaps in the same. For the defined research gaps we develop novel techniques that address relevance of features to the target and redundancy of features amidst themselves. Our techniques aim at handling homogeneous data, i.e., only continuous or categorical features, mixed data, i.e., continuous and categorical features, and time series
Subjects
Feature-Selektion, multivariate Korrelationsanalyse, Feature-Redundanz, mehrdimensionalen Wechselwirkungen, erklärbare Korrelationen, Ordinalmuster, kontinuierliche und kategorische Daten, Zeitreihenanalyse, filterbasierte Ansätze, feature selection, higher-order interactions, redundancy, filter-based approach, ordinal patterns, feature extraction, explainable correlations
Classification (DDC)
004 Informatik
Shekar, Arvind Kumar: Multivariate Correlation Analysis for Supervised Feature Selection in High-Dimensional Data. - Bonn, 2020. - Dissertation, Rheinische Friedrich-Wilhelms-Universität Bonn.
Online-Ausgabe in bonndoc: https://nbn-resolving.org/urn:nbn:de:hbz:5-58003
@phdthesis{handle:20.500.11811/8303,
urn: https://nbn-resolving.org/urn:nbn:de:hbz:5-58003,
author = {{Arvind Kumar Shekar}},
title = {Multivariate Correlation Analysis for Supervised Feature Selection in High-Dimensional Data},
school = {Rheinische Friedrich-Wilhelms-Universität Bonn},
year = 2020,
month = mar,
note = {The main theme of this dissertation focuses on multivariate correlation analysis on different data types and we identify and define various research gaps in the same. For the defined research gaps we develop novel techniques that address relevance of features to the target and redundancy of features amidst themselves. Our techniques aim at handling homogeneous data, i.e., only continuous or categorical features, mixed data, i.e., continuous and categorical features, and time series},
url = {https://hdl.handle.net/20.500.11811/8303}
}

The following license files are associated with this item:

InCopyright