Despite the development of many dynamic multi-objective evolutionary algorithms (DMOEAs), most still struggle to comprehensively address various types of dynamic multi-objective optimization problems (DMOPs). To address this issue, this paper proposes a dynamic multi-objective optimization algorithm based on adaptive response selection (ARS-DMOEA). The core idea of the ARS-DMOEA is to adaptively select dynamic response strategies with varying strengths to effectively handle various types of DMOPs. Firstly, an adaptive response selection strategy is introduced. This strategy can dynamically adjust the selection probabilities of different dynamic response strategies based on their historical performance. Then, a hybrid dynamic response strategy is designed. It selects individuals generated by different strategies according to their selection probabilities, thus creating a high-quality initial population in new environments. Comparative experiments are conducted against four state-of-the-art DMOEAs. The results indicate that the ARS-DMOEA not only demonstrates significant competitiveness but also effectively adapts to various types of dynamic multi-objective optimization problems.