九州大学-研究者情報 [廣瀨慧 (教授) マス・フォア・インダストリ研究所産業数理統計研究部門]

廣瀨慧（ひろせけい）

データ更新日：2023.11.22

教授　／　マス・フォア・インダストリ研究所産業数理統計研究部門

原著論文

1.	Kei Hirose, Kanta Miura, Atori Koie, Hierarchical clustered multiclass discriminant analysis via cross-validation, Computational Statistics and Data Analysis, 10.1016/j.csda.2022.107613, 178, 2023.02, Linear discriminant analysis (LDA) is a well-known method for multiclass classification and dimensionality reduction. However, in general, ordinary LDA does not achieve high prediction accuracy when observations in some classes are difficult to be classified. A novel cluster-based LDA method is proposed that significantly improves prediction accuracy. Hierarchical clustering is adopted, and the dissimilarity measure of two clusters is defined by the cross-validation (CV) value. Therefore, clusters are constructed such that the misclassification error rate is minimized. The proposed approach involves a heavy computational load because the CV value must be computed at each step of the hierarchical clustering algorithm. To address this issue, a regression formulation for LDA is developed and an efficient algorithm that computes an approximate CV value is constructed. The performance of the proposed method is investigated by applying it to both artificial and real datasets. The proposed method provides high prediction accuracy with fast computation from both numerical and theoretical viewpoints..
2.	Kei Hirose, Interpretable Modeling for Short- and Medium-Term Electricity Demand Forecasting, FRONTIERS IN ENERGY RESEARCH, 10.3389/fenrg.2021.724780, 9, 2021.12, We consider the problem of short- and medium-term electricity demand forecasting by using past demand and daily weather forecast information. Conventionally, many researchers have directly applied regression analysis. However, interpreting the effect of weather on the demand is difficult with the existing methods. In this study, we build a statistical model that resolves this interpretation issue. A varying coefficient model with basis expansion is used to capture the nonlinear structure of the weather effect. This approach results in an interpretable model when the regression coefficients are nonnegative. To estimate the nonnegative regression coefficients, we employ nonnegative least squares. Three real data analyses show the practicality of our proposed statistical modeling. Two of them demonstrate good forecast accuracy and interpretability of our proposed method. In the third example, we investigate the effect of COVID-19 on electricity demand. The interpretation would help make strategies for energy-saving interventions and demand response..
3.	Keisuke Teramoto, Kei Hirose, Sparse multivariate regression with missing values and its application to the prediction of material properties, INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN ENGINEERING, 10.1002/nme.6867, 123, 2, 530-546, 2022.01, In the field of materials science and engineering, statistical analysis and machine learning techniques have recently been used to predict multiple material properties from an experimental design. These material properties correspond to response variables in the multivariate regression model. In this study, we conduct a penalized maximum likelihood procedure to estimate model parameters, including the regression coefficients and covariance matrix of response variables. In particular, we employ l1-regularization to achieve a sparse estimation of The regression coefficients and inverse covariance matrix of response variables. In some cases, there may be a relatively large number of missing values in the response variables, owing to the difficulty of collecting data on material properties. We therefore propose a method that incorporates a correlation structure among the response variables into a statistical model to improve the prediction accuracy under the situation with missing values. The expectation maximization algorithm is also constructed, which enables application to a dataset with missing values in the responses. We apply our proposed procedure to real data consisting of 22 material properties..
4.	Kei Hirose, Yoshikazu Terada, Sparse and Simple Structure Estimation via Prenet Penalization, PSYCHOMETRIKA, 10.1007/s11336-022-09868-4, 2022.05, We propose a prenet (product-based elastic net), a novel penalization method for factor analysis models. The penalty is based on the product of a pair of elements in each row of the loading matrix. The prenet not only shrinks some of the factor loadings toward exactly zero but also enhances the simplicity of the loading matrix, which plays an important role in the interpretation of the common factors. In particular, with a large amount of prenet penalization, the estimated loading matrix possesses a perfect simple structure, which is known as a desirable structure in terms of the simplicity of the loading matrix. Furthermore, the perfect simple structure estimation via the proposed penalization turns out to be a generalization of the k-means clustering of variables. On the other hand, a mild amount of the penalization approximates a loading matrix estimated by the quartimin rotation, one of the most commonly used oblique rotation techniques. Simulation studies compare the performance of our proposed penalization with that of existing methods under a variety of settings. The usefulness of the perfect simple structure estimation via our proposed procedure is presented through various real data applications..
5.	Okinaga Y, Kyogoku D, Kondo S, Nagano A, Hirose K, Event Effects Estimation on Electricity Demand Forecasting., Scientific Reports, 11, 2021.06, [URL].
6.	Hirose K, Wada K, Hori M,Taniguchi R, Relationship between gene regulation network structure and prediction accuracy in high dimensional regression., Energies, 13, 21, 2020.11, [URL], 遺伝子ネットワークと予測精度をシミュレーションし，そのソフトウェアを公開した．.
7.	廣瀬慧, L1正則化法に基づく因子分析および構造方程式モデリングの最近の展開, 計算機統計学会, （採録決定）, 2020.06.
8.	Kei Hirose, Sparse modeling and model selection, Journal of the Institute of Electronics, Information and Communication Engineers, 99, 5, 392-399, 2016.05.
9.	Miyuki Imada, Kei Hirose, Manabu Yoshida, Sun Yong Kim, Naoya Toyozumi, Guillaume Lopez, Yutaka Kano, An interpersonal sentiment quantification method applied to work relationship prediction, NTT Technical Review, 15, 3, 2017.03, For a business to be successful, it is important for people in the business to consider how other people feel, that is, to consider interpersonal sentiment. Our research goal is to quantitatively predict the strength of interpersonal sentiment by analyzing a small amount of data on office employees, for example, their gender or age group, and data on events such as giving positive feedback on work done and sexual or power harassment without directly asking someone about their change in sentiment. In this article, we propose an interpersonal-sentiment-changing model for this quantification and propose two new analysis methods for developing prediction formulas. These methods can be used even if 90% of data is missing and in environments in which it is difficult to gather data in a comparatively short time. We also implement two visualization systems to predict how interpersonal sentiment changes for each event based on actual office data..
10.	Kei Hirose, Hironori Fujisawa, Jun Sese, Robust sparse Gaussian graphical modeling, Journal of Multivariate Analysis, 10.1016/j.jmva.2017.07.012, 161, 172-190, 2017.09, [URL], Gaussian graphical modeling is popular as a means of exploring network structures, such as gene regulatory networks and social networks. An L₁ penalized maximum likelihood approach is often used to learn high-dimensional graphical models. However, the penalized maximum likelihood procedure is sensitive to outliers. To overcome this problem, we introduce a robust estimation procedure based on the γ-divergence. The proposed method has a redescending property, which is a desirable feature in robust statistics. The parameter estimation procedure is constructed using the Majorize-Minimization algorithm, which guarantees that the objective function monotonically decreases at each iteration. Extensive simulation studies show that our procedure performs much better than the existing methods, in particular, when the contamination ratio is large. Two real data analyses are used for illustration purposes..
11.	Ján Dolinský, Kei Hirose, Sadanori Konishi, Readouts for echo-state networks built using locally regularized orthogonal forward regression, Journal of Applied Statistics, 10.1080/02664763.2017.1305331, 45, 4, 740-762, 2018.03, [URL], Echo state network (ESN) is viewed as a temporal expansion which naturally give rise to regressors of various relevance to a teacher output. We illustrate that often only a certain amount of the generated echo-regressors effectively explain the teacher output and we propose to determine the importance of the echo-regressors by a joint calculation of the individual variance contributions and Bayesian relevance using the locally regularized orthogonal forward regression (LROFR). This information can be advantageously used in a variety of ways for an analysis of an ESN structure. We present a locally regularized linear readout built using LROFR. The readout may have a smaller dimensionality than the ESN model itself, and improves robustness and accuracy of an ESN. Its main advantage is ability to determine what type of an additional readout is suitable for a task at hand. Comparison with PCA is provided too. We also propose a radial basis function (RBF) readout built using LROFR, since flexibility of the linear readout has limitations and might be insufficient for complex tasks. Its excellent generalization abilities make it a viable alternative to feed-forward neural networks or relevance-vector-machines. For cases where more temporal capacity is required we propose well studied delay&sum readout..
12.	Kei Hirose, Hiroki Masuda, Robust relative error estimation, Entropy, 10.3390/e20090632, 20, 9, 2018.08, [URL], Relative error estimation has been recently used in regression analysis. A crucial issue of the existing relative error estimation procedures is that they are sensitive to outliers. To address this issue, we employ the γ-likelihood function, which is constructed through γ-cross entropy with keeping the original statistical model in use. The estimating equation has a redescending property, a desirable property in robust statistics, for a broad class of noise distributions. To find a minimizer of the negative γ-likelihood function, a majorize-minimization (MM) algorithm is constructed. The proposed algorithm is guaranteed to decrease the negative γ-likelihood function at each iteration. We also derive asymptotic normality of the corresponding estimator together with a simple consistent estimator of the asymptotic covariance matrix, so that we can readily construct approximate confidence sets. Monte Carlo simulation is conducted to investigate the effectiveness of the proposed procedure. Real data analysis illustrates the usefulness of our proposed procedure..
13.	Yamamoto, M., Hirose, K., Nagata, H., Graphical tool of sparse factor analysis, Behaviormetrika, Volume 44, Issue 1, 2017.01.
14.	Hirose, K. and Imada, M., Sparse factor regression via penalized maximum likelihood estimation., Statistical Papers, 10.1007/s00362-016-0781-8, 633-662, 2018.05, In factor regression model, the maximum likelihood estimation suffers from three disadvantages: (i) the maximum likelihood estimates are unavailable when the number of variables exceeds the number of observations, (ii) the rotation technique based on maximum likelihood estimates produces an insufficiently sparse loading matrix, and (iii) multicollinearity can occur when the estimates of unique variances (specific variances) are small because the regression coefficients are sensitive to the inverse of unique variances. To handle these problems, we propose a penalized maximum likelihood procedure. Specifically, we impose a lasso-type penalty on the factor loadings to improve the sparseness of the solution. We also introduce a penalty on unique variances, which (given the factor scores) corresponds to the ridge penalty on the regression coefficient. Theoretical properties from a prediction viewpoint of our procedure are discussed. The effectiveness of the procedure is investigated through Monte Carlo simulations. The utility of our procedure is demonstrated on real data collected by an online questionnaire..

本データベースの内容を無断転載することを禁止します。

九大関連コンテンツ

pure2017年10月2日から、「九州大学研究者情報」を補完するデータベースとして、Elsevier社の「Pure」による研究業績の公開を開始しました。

　

QIR　九州大学学術情報リポジトリシステム情報科学研究院

システム情報科学研究院