〇劉海龍，谷口忠大，高野敏明，竹中一仁，坂東誉司，田中雄介，Deep Sparse Autoencoderによる車両運転状態の可視化，2014年度人工知能学会全国大会 (JSAI2014), 3H3-OS-24a-3in, (2014)

发布日期：May 14, 2014 9:19:48 AM

Abstract:

With the increase of dimensions in driving behavioral data, the human to intuitively understand the timeseries data has become very difficult. We employed a deep sparse autoencoder to extract the three-dimensional representation from raw driving behavioral data with one hundred dimensions. And we proposed a method to visualize driving behavioral on the map by mapping three-dimensional data into the RGB color space. We compared deep sparse autoencoder with other conventional methods such as principal component analysis. As a result, our methods outperformed other conventional methods for visualization of driving behavioral data.

人工知能学会第28回全国大会優秀賞