A Top_kws Algorithm for Synthetics and Real Datasets

Abstract

Recommendation Systems (RS) is the most commonly used Information technology in the last decade. RS’s processing behavior can be found on different approaches, such as Decision Making (DM) support and preferences query processing. These systems have been used in many Internet activities, mainly to overcome information overload and for many other purposes. Some of these include e-commerce sites, web page searching, elearning, and Cloud Computing Services. Also, research has been conducted on the use of RS in some sport management Systems. This paper presents a performance evaluation of the Topkws recommendation algorithm, applying on a new RS called Generic Research and Selection System (GRSS) based on the Skyline and the Topk query processing. The conceived algorithm, which used an adapted Multi-criteria Decision Aiding (MCDA) method, was applied to different research eras in this paper, namely the Cloud Computing Services and Sport management System. The algorithm shows to be more and more efficient. Extensive experiments based on correlation study toward both real and synthetic datasets demonstrate the efficiency and scalability of this algorithm compared with other best-known algorithms in this field.