How distance metrics influence missing data imputation with k-nearest neighbours

• Type: Article in International Scientific Journal
• Date: 2020

Authors

Santos, MS (Author) Other; Pedro Henriques Abreu (Author) Other; Wilk, S (Author) Other; Santos, J (Author) Other

Journal

Title: Pattern Recognition Letters

Vol. 136

Pages: 111-119

ISSN: 0167-8655

Publisher: Elsevier

Indexing

ISI Web of Knowledge - 30 Citations

Scopus - 35 Citations

Other information

Authenticus ID: P-00S-9GF

DOI: 10.1016/j.patrec.2020.05.032

Abstract (EN): In missing data contexts, k-nearest neighbours imputation has proven beneficial since it takes advantage of the similarity between patterns to replace missing values. When dealing with heterogeneous data, researchers traditionally apply the HEOM distance, that handles continuous, nominal and missing data. Although other heterogeneous distances have been proposed, they have not yet been investigated and compared for k-nearest neighbours imputation. In this work, we study the effect of several heterogeneous distances on k-nearest neighbours imputation on a large benchmark of publicly-available datasets.

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 9