ISOIS, an instrument suite onboard NASA’s Parker Solar Probe, is helping scientists to study energetic particles released by the Sun during solar storms. ISOIS measures the direction and intensity of these particles, which is important for understanding how they travel through space. However, the instrument sometimes misses data when it’s not aligned with the local magnetic field. This creates gaps in the measurements that are difficult to fill using traditional methods. In this study, we use machine learning models to predict the missing data during Solar Energetic Particle (SEP) events and create a more complete picture of how these particles move. Approximately half of the solar energetic particle (SEP) event samples studied in this work show strong, reliable reconstructions of missing data. This allows us to correct false drops in cumulative intensity caused by missing observations. Since this type of data loss occurs on many space missions that use fixed instruments, our method can be applied more broadly to help fill data gaps and improve the quality of space weather observations.