Analysis and Evaluation of a Framework for Sampling Database in Recommenders

Abstract

In this paper the authors proposed a database sampling framework that aims to minimize the time necessary to produce a sample database. They argue that the performance of current relational database sampling techniques that maintain the data integrity of the sample database is low and a faster strategy needs to be devised. The sampling method targets the production environment of a system under development that generally consists of large amounts of data computationally costly to analyze. The results have been improved due to the fact that the authors have selected the users that they had more information about them and they have made the data table denser. Therefore, by increasing the data and making the rating more comprehensive for all the users they can help to produce the more and better association rules. The obtained results were not that much suitable for Jester dataset but with their proposed methods the authors have tried to improve the quantity and quality of the rules. These results indicate that the effectiveness of the system greatly depends on the input data and the applied dataset. In addition, if the user rates more number of the items the system efficiency will be more increased.