Abstract
A system of structuring images gathered by Google image search API and the neighbor texts using Formal Concept Analysis (FCA), is developed. Due to the variety of texts, it is difficult to do the clustering the images based on similarities. In order to solve the problem, we propose a method of adding adjectives to images based on classifiers. The classifiers are constructed through the learning process using training data which includes images and corresponding adjectives. We perform experiments using Google image retrieval system and confirm the effectiveness of the proposed method.
