|Ryuei Nishii||Last modified date：2018.10.23|
My research topics are shown here, and messages related to lectures are also available. .
Doctor of Science
Field of Specialization
Owing to the development of sensor technology, a huge amount of digital data has been acquired in the various fields. Then, effective data analysis methods for elucidating important information from huge data are required. Traditional techniques, e.g. statistical approach are, however, not based on the situation. Pattern recognition due to machine learning is one of techniques for huge data analysis, and it has been progressed since the method shows high performance for massive data like imagery and movies. My research interest is to propose new statistical machine learning techniques for data analysis, and to evaluate them through real data such as remotely sensed imagery and/or genome data.
Research InterestsMembership in Academic Society
- Statistical modeling of spatio-temporal phenomena
keyword : model selection, pattern recognition, image classification
- Image classification and unmixing based on multi-spectral imagery
keyword : remotely-sensed imagery, learning theory, Markov random fields
2002.04～2014.03Development of mathematical methodology for discrimination of land cover/usage based on multispectral imagery aquired by aritificial satellites.
- Supervised image classification based on machine learning techniques
|1.||Koda,S., Onda,Y., Matsui,H., Takahagi,K., Uehara-Yamaguchi,Y., Shimizu,M., Inoue,K., Yoshida,T., Sakurai,T., Honda,H.,Eguchi,S.,Nishii,R. and Mochida,K., Diurnal transcriptome and gene network represented through sparse modeling in brachypodium distachyon, Frontiers in Plant Science, 10.3389/fpls.2017.02055, 2017.11, [URL], 試験植物ブラキポディウムの4時間ごとに計測した遺伝子発現量の時系列データに対し，時間遅れを伴う他の遺伝子発現量および自己回帰項を含む回帰モデルを想定した．母数推定にはグループSCADによるスパース推定，および安定的なネットワーク推定を行い，ハブ遺伝子を検出した．これにより遺伝子間ネットワークができた．.|
|2.||Ryuei Nishii, Shojiro Tanaka, Modeling and inference of forest coverage ratio using zero-one inflated distributions with spatial dependence., Environmental and Ecological Statistics, 10.1007/s10651-012-0227-y, 2013.06, This paper explores statistical modeling of forest area with two covariates. The forest coverage ratio of grid-cell data was modeled by taking human population density and relief energy into account. The likelihood of the forest ratios was decomposed into the product of two likelihoods. The first likelihood was due to trinomial logistic distributions on three categories: the forest ratios take zero, or one, or values between zero and one. The second one was due to a logistic-normal regression model for the ratios between zero and one. This model was applied to real grid-cell data and resulted in remarkably interesting implications..|
|3.||R. Nishii and S. Eguchi, Supervised image classification by contextual AdaBoost based on posteriors in neighborhoods, IEEE Transactions on Geoscience and Remote Sensing, 10.1109/TGRS.2005.848693, 43, 11, 2547-2554, Vol. 43 (11), 2547-2554, 2005.01.|
|4.||R. Nishii, A Markov random field-based approach to decision level fusion for remote sensing image classification, IEEE Transactions on Geoscience and Remote Sensing, 41(10) pp. 2316-2319, 2003.10.|
|5.||R. Nishii, S. Kusanobu and N. Nakaoka, Selection of variables and neighborhoods for spatial enhancement of thermal infrared images, Communication in Statistics -Theory and Methods, 28(3 & 4), pp. 965-976, 1999.01.|
|6.||R. Nishii, S. Kusanobu and S. Tanaka, Enhancement of low spatial resolution image based on high resolution bands, IEEE Transactions on Geoscience and Remote Sensing, 34(5), pp. 1151-1158, 1996.01.|
|1.||Statistical modeling of hyper-dimensional data and spatio-temporal data, [URL].|
|2.||Ryuei Nishii, Statistical deforestation modeling based on zero-one inflated distributions with spatial dependence
, Forum for Interdisciplinary Mathematics 2010, 2010.12, [URL].
|3.||Ryuei Nishii, Semi-Supervised Contextual Unmixing of Geospatial Data
, IASC 2008, 2008.12.
|4.||Applications of statistical methods to remotely-sensed imagery.|
- Asia-Pacific Consortium of Mathematics for Industry (APCMfI)
- The paper "Markov random field based on Kullback-Leibler divergence and its applications to geo-spatial image segmentation" presented by R. Nishii was elected as one of best papers presented in the conference.
I give lectures on introduction of statistics and mathematical statistics in undergraduate courses, and on multivariate statistical analysis and on pattern recognition based on statistical machine learning in graduate courses. Similar lectures are also given in intensive courses of other universities. Further, I study mathematical statistics and statistical machine learning through actual data with my seminarians.