[2K5-GS-2] Machine learning

[2K5-GS-2-05] A Study of the Effect of Season on Maximum Temperature Prediction Using Transfer Learning

〇HIDEAKI DOBASHI¹, SHOICHI URANO¹ (1. Meiji University)

Keywords:Transfer learning, Machine learning, Time series data, Maximum temperature forecast

The amount of data handled in neural networks using machine learning is an important factor that affects accuracy. However, not all models have a sufficient amount of data. Therefore, I focused on transition learning, one of the machine learning methods.This study examines which periods a given target data set is valid for by testing whether a model trained on a small amount of data using transition learning predicts maximum temperatures with high accuracy in any given month.