Diversity among base classifiers plays an important role in ensemble learning for illustrating the working mechanism of multiple classifier systems and constructing effective ensemble systems. However, at present, there doesn’t exist a widely accepted diversity measure. Firstly, some commonly used diversity measures are summarized, and the perspective adopted by each measure is illustrated when evaluating the diversity. Then, the research progresses of diversity
measures are investigated in the following three aspects: The recently proposed interpretations and measures for diversity, the application of diversity measures in selective ensemble, and the relationship between diversity measures and ensemble accuracy. Finally, several directions for future research are given.