In this study, the PM2.5, prediction performances of long short-term memory (LSTM), gated recurrentunit (GRU), and bidirectional LSTM (Bi-LSTM) were compared using data from Seoul, Daejeon, andBusan, which are representative cities in Korea. The data analysis period was from 9:00 on May 16,2014, to 23:00 on December 31, 2021, based on data at 1 h intervals. The causal factors affecting thechange in PM2.5 of three cities in Korea, and five major cities in China were determined. The analysisrevealed that the three models showed similarly high performances in short-term prediction within24 h (R2 0.9). The Bi-LSTM model using both past and future time information showed high predictionaccuracy for long-term prediction (R2 0.6). Using the PM2.5 data of the five major Chinese cities, it wasconfirmed that the accuracy of the PM2.5 prediction model for Seoul, Daejeon, and Busan improved. Thedeep learning model showed a high accuracy even when the Fine Dust Act measures were implemented. This study can facilitate governments to prepare measures against air pollution with a high regional predictionperformance by identifying the causal factors affecting PM2.5, specific to the city, and designingdifferent models for each forecasting period.