Since online images may not accurately reflect the actual weather conditions at a destination, tourists often find themselves in unexpected scenarios. To address this, we leverage CycleGAN for seasonal style transfer, generating images that depict locations across all four seasons. Additionally, to streamline itinerary planning, we develop an Autoencoder-based recommender system that personalizes travel suggestions.