Study on a system for adding a new utility value on Kana calligraphy images

• Project/Area Number: 23650091
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Perception information processing/Intelligent robotics
• Research Institution: The Institute of Statistical Mathematics
• Principal Investigator: MATSUI Tomoko 統計数理研究所, モデリング研究系, 教授 (10370090)
• Project Period (FY): 2011 – 2013
• Project Status: Completed (Fiscal Year 2013)
• Budget Amount: ¥3,640,000 (Direct Cost: ¥2,800,000、Indirect Cost: ¥840,000); Fiscal Year 2013: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000); Fiscal Year 2012: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000); Fiscal Year 2011: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
• Keywords: 機械学習 / サポートベクターマシン / かな書データ / 体調判定 / 統計科学 / かな書

Research Abstract

I studied two topics, (1) creation of a Kana calligraphy image database and (2) construction of a system for predicting the mood and physical conditions from Kana calligraphy images. For (1), Kana calligraphy images and survey responses on the mood and physical conditions by 5 subjects were collected once a week over 8 months. For (2), support vector machines with polynomial kernels were used to predict the mood and physical conditions from the Kana calligraphy images. It was found that especially for the body temperature and the eye strains, the average precisions were relatively high.