Abstract Bandit algorithms have been widely used in many application areas including information retrieval evaluation and ranking. This is largely due to their exceptional performance. The aim of this study is to examine the overall published studies in terms of trends that shape the use of bandit algorithms in the evaluation and ranking of information retrieval systems. This study also seeks to classify the bandit algorithms used in the research domain. In totality the evaluation metrics, datasets, contribution facets of primary studies as well as the bandit categories are discussed.