In this project, we aim to develop methods to extract “quality” of information from images that are related to human subjective judgements and to select images that have high “value” from a very number of images. As an example, we constructed a deep learning model to learn and predict several different subjective judgments made by human (desire to eat, whether it is made for young people, etc.) for food images. We are also investigating the image features that contributed to human judgement by analyzing the learned model. The results of these researches provide insights about how humans use image features to make subjective evaluations.