M-RS: A Systematic Literature Review on Multi-Objective Optimization along with Recommendation System
Muteeb Bin Muzammil, Sahar Ajmal, Shahzeb Javed, Arsalan Tariq, Umer Iqbal · 2021 International Conference on Engineering and Emerging Technologies (ICEET) · 2021
Recommendation systems (RS) are used for extracting data on the basis of rating and interest provided by the users on a particular item. In the current era, the world is relying more on RS to get the relevant data. To increase the efficiency of RS, Multi-Objective Optimization (MOO) is integrated with it. MOO is a branch of a multi-criteria decision-making scheme. MOO considers different conflicting objectives to optimize the results. The combination of MOO and RS will enhance the reliability of recommendations because the recommendations are provided by using different intentions of users with diverse requirements. This paper performs a Systematic Literature Review (SLR) on the articles that use MOO and RS with each other. The SLR is performed on 26 studies that are extracted from the study selection procedure. These 26 articles are then analyzed through different perspectives such as their domains, the method they have used, and evaluation techniques adopted, and the overall research trend of M-RS. The result shows that the application of MOO in field of RS is most popular among generic RS rather than specific RS. Furthermore, the results reflect that the MovieLens dataset has the highest popularity as 13 studies used it for the evaluation of their techniques. In the future, more research questions will be proposed to have a more in-depth insight into this field.