Rainfall plays a critical role in the global water and energy cycle, influencing surface water availability and recharge processes both spatially and temporally. Traditional rainfall data collection using ombrometers provides accurate live data, but often faces the challenge of missing data due to equipment failure or transmission, especially in agencies such as BMKG. This study analyzes the effectiveness of Artificial Neural Network (ANN) and Inverse Distance Weighting (IDW) methods in imputing missing rainfall data in Semarang City, using data from 31 observation stations over a 34-year period. Of the total 177,093 data, there are 7,159 data that require imputation. The ANN model showed superior performance with an RMSE of 1.2231 mm and R² of 0.9961 in the wet season, and an RMSE of 0.9489 mm and R² of 0.9926 in the dry season. Meanwhile, the IDW method showed limitations with an RMSE of 18.8206 mm and R² of 0.0084 in the wet season, and an RMSE of 10.9974 mm and R² of 0.0019 in the dry season. Although IDW achieved perfect recall (1.0000) in rainfall event classification, its low precision resulted in a suboptimal F1-Score (0.6460 wet season; 0.3124 dry season). The results indicate that ANN is superior in capturing non-linear patterns of rainfall and adapting to seasonal variations, while IDW has significant limitations in explaining rainfall variability. These findings make an important contribution to the development of more reliable rainfall data imputation methods for climatology and hydrology applications in urban areas.