Abstract

Prediction problems have been highly studied in the literature especially in service systems. Recently machine learning based solutions have been added to find optimal solutions to such problems. In this study, the aim is to compile the existing recent literature about prediction problems in various fields. A systematic literature review process has been followed, using ScienceDirect, Scopus and Web of Science. After filtering the outcomes of the databases 24 articles are identified to be fully analyzed. The results indicate that formulating the prediction as a classifier problem, which enables usage of XGBoost, Bayesian Classifier, regression etc. provides need for labelled data, and provide computational ease, whereas can be less accurate, compared to deep learning solutions. Deep learning solutions are, however, “black-box” as in in any field, literature cannot fully identify each iteration that has been recorded. Finally, literature suggests combining into hybrid solutions, as in using non-ML and ML solutions together, for example heuristic rulesets and deep learning to obtain a better fit.

Download

Share

COinS