Academic Journals Database
Disseminating quality controlled scientific knowledge

Efficient Formulations for 1-SVM and their Application to Recommendation Tasks

Author(s): Yasutoshi Yajima | Tien-Fang Kuo

Journal: Journal of Computers
ISSN 1796-203X

Volume: 1;
Issue: 3;
Start page: 27;
Date: 2006;
Original page

Keywords: support vector machine | Laplacian matrix | graph kernel | quadratic programming problem | collaborative filtering | recommender system

The present paper proposes new approaches for recommendation tasks based on one-class support vector machines (1-SVMs) with graph kernels generated from a Laplacian matrix. We introduce new formulations for the 1-SVM that can manipulate graph kernels quite efficiently. We demonstrate that the proposed formulations fully utilize the sparse structure of the Laplacian matrix, which enables the proposed approaches to be applied to recommendation tasks having a large number of customers and products in practical computational times. Results of various numericalexperiments demonstrating the high performance of the proposed approaches are presented.

Tango Rapperswil
Tango Rapperswil

     Affiliate Program