Far-Infrared (FIR) observation is important for studying galaxies and star-formation(SF), especially the FIR plays an important role in estimating the star-formation rate(SFR) in spectral energy distribution (SED) fitting because the galaxies’ dust re-emits radiation in the FIR. However, there are no available telescopes to perform observations in the FIR flux yet, and so many galaxies do not have FIR measurements, in this study, we adopt the data in AKARI to predict the FIR flux since machine learning (ML) is powerful and widely used in data analysis. We use two different models, support vector machine (SVM) and neural network (NN). In the ML model, we input 23 bands fluxes from ultraviolet to mid-infrared flux to predict the FIR flux. Via the ML techniques, we can predict the FIR flux for galaxies and figure out the features between multiwavelength fluxes.