446 0

l-Injection: Toward Effective Collaborative Filtering Using Uninteresting Items

Title
l-Injection: Toward Effective Collaborative Filtering Using Uninteresting Items
Author
김상욱
Keywords
Recommender systems; collaborative filtering; data sparsity; uninteresting items; pre-use preference; post-use preference
Issue Date
2019-01
Publisher
IEEE COMPUTER SOC
Citation
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, v. 31, NO 1, Page. 3-16
Abstract
We develop a novel framework, named as iota-injection, to address the sparsity problem of recommender systems. By carefully injecting low values to a selected set of unrated user-item pairs in a user-item matrix, we demonstrate that top-N recommendation accuracies of various collaborative filtering (CF) techniques can be significantly and consistently improved. We first adopt the notion of pre-use preferences of users toward a vast amount of unrated items. Using this notion, we identify uninteresting items that have not been rated yet but are likely to receive low ratings from users, and selectively impute them as low values. As our proposed approach is method-agnostic, it can be easily applied to a variety of CF algorithms. Through comprehensive experiments with three real-life datasets (e.g., Movielens, Ciao, and Watcha), we demonstrate that our solution consistently and universally enhances the accuracies of existing CF algorithms (e.g., item-based CF, SVD-based CF, and SVD++) by 2.5 to 5 times on average. Furthermore, our solution improves the running time of those CF methods by 1.2 to 2.3 times when its setting produces the best accuracy.
URI
https://ieeexplore.ieee.org/document/7913668https://repository.hanyang.ac.kr/handle/20.500.11754/132980
ISSN
1041-4347; 1558-2191
DOI
10.1109/TKDE.2017.2698461
Appears in Collections:
COLLEGE OF ENGINEERING[S](공과대학) > COMPUTER SCIENCE(컴퓨터소프트웨어학부) > Articles
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE