The Regression Tree Model Building Based on a Cluster-Regression Approximation for Data-Driven Medicine

被引：0

作者：

Subbotin, Sergey ^{[1
]}

Kirsanova, Elena ^{[2
]}

机构：

[1] Zaporizhzhia Natl Tech Univ, Zhukovsky Str 64, UA-69063 Zaporizhzhia, Ukraine

[2] Zaporizhzhia State Med Univ, Maiakovskyi Ave 26, UA-69035 Zaporizhzhia, Ukraine

来源：

PROCEEDINGS OF THE 1ST INTERNATIONAL WORKSHOP ON INFORMATICS & DATA- DRIVEN MEDICINE (IDDM 2018) | 2018年 / 2255卷

关键词：

data-driven diagnosis; regression; cluster analysis; cluster-regression approximation; regression tree; neural network; NEURAL-NETWORKS;

D O I：

暂无

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

The problem of quantitative dependency model building on precedents for data-driven medicine is considered. A tree-cluster-regression approximation method is proposed. It makes possible to ensure acceptable model accuracy, high levels of interpretability and generalization of data, and to reduce the complexity of the model. The software that implements the proposed methods is developed. The developed software is studied at solving the problem of children health indicator modelling.

引用

页码：155 / 169

页数：15

共 31 条

[1]

Abonyi J., 2007, CLUSTER ANAL DATA MI

[2]

AFIFI A.A., 1979, STAT ANAL COMPUTER O

[3] NEW LOOK AT STATISTICAL-MODEL IDENTIFICATION [J].

AKAIKE, H .

IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1974, AC19 (06) :716-723

[4]

[Anonymous], 2019, Inductive learning algorithms for complex systems modeling

[5]

[Anonymous], 1997, IEEE T AUTOM CONTROL, DOI DOI 10.1109/TAC.1997.633847

[6]

[Anonymous], 2009, Encyclopedia of Complexity and Systems Science, DOI DOI 10.1007/978-0-387-30440-3_301

[7]

Breiman L., 1983, OLSHEN STONE CLASSIF

[8] FUZZY NEURAL NETWORKS - A SURVEY [J].

BUCKLEY, JJ ;

HAYASHI, Y .

FUZZY SETS AND SYSTEMS, 1994, 66 (01) :1-13

[9]

Clarke B, 2009, SPRINGER SER STAT, P1, DOI 10.1007/978-0-387-98135-2

[10]

Gorban A. N., 1999, IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339), P4393, DOI 10.1109/IJCNN.1999.830876

← 1 2 3 4 →