Associative classifiers (ACs) have been used in several areas due to its ability to generate interpretable models. Although ACs present good results, when applied to imbalanced problems, performance does not remain the same. Standard classifiers are generally biased towards the majority class in favor of measuring accuracy. Thus, specific rules aimed at the minority class end up being ignored, causing instances of this class to be incorrectly classified. Therefore, solutions are being developed. This work presents a systematic literature review on algorithm-level solutions aimed at ACs for imbalanced data. The aim is (i) to provide a better understanding of the strategies used in each of the models induction steps in order to treat imbalance, (ii) identify gaps and opportunities in the area, (iii) support the development and/or expansion of packages aimed at ACs algorithms for imbalanced data.