Self-organizing linear output map (SOLO): An artificial neural network suitable for hydrologic modeling and analysis

Kuo Lin Hsu, Hoshin Vijai Gupta, Xiaogang Gao, Soroosh Sorooshian, Bisher Imam

Research output: Contribution to journalArticle

147 Citations (Scopus)

Abstract

Artificial neural networks (ANNs) can be useful in the prediction of hydrologic variables, such as streamflow, particularly when the underlying processes have complex nonlinear interrelationships. However, conventional ANN structures suffer from network training issues that significantly limit their widespread application. This paper presents a multivariate ANN procedure entitled self-organizing linear output map (SOLO), whose structure has been designed for rapid, precise, and inexpensive estimation of network structure/parameters and system outputs. More important, SOLO provides features that facilitate insight into the underlying processes, thereby extending its usefulness beyond forecast applications as a tool for scientific investigations. These characteristics are demonstrated using a classic rainfall-runoff forecasting problem. Various aspects of model performance are evaluated in comparison with other commonly used modeling approaches, including multilayer feedforward ANNs, linear time series modeling, and conceptual rainfall-runoff modeling.

Original languageEnglish (US)
Pages (from-to)381-3817
Number of pages3437
JournalWater Resources Research
Volume38
Issue number12
StatePublished - Dec 1 2002

Fingerprint

artificial neural network
neural networks
Neural networks
Runoff
modeling
Rain
runoff
hydrologic factors
rain
rainfall-runoff modeling
stream flow
streamflow
Time series
time series analysis
Multilayers
time series
rainfall
prediction
analysis

Keywords

  • Artificial neural network
  • Overfitting
  • Principal component analysis
  • Rainfall-runoff modeling
  • Self-organizing feature map
  • SOLO

ASJC Scopus subject areas

  • Environmental Science(all)
  • Environmental Chemistry
  • Aquatic Science
  • Water Science and Technology

Cite this

Self-organizing linear output map (SOLO) : An artificial neural network suitable for hydrologic modeling and analysis. / Hsu, Kuo Lin; Gupta, Hoshin Vijai; Gao, Xiaogang; Sorooshian, Soroosh; Imam, Bisher.

In: Water Resources Research, Vol. 38, No. 12, 01.12.2002, p. 381-3817.

Research output: Contribution to journalArticle

Hsu, Kuo Lin ; Gupta, Hoshin Vijai ; Gao, Xiaogang ; Sorooshian, Soroosh ; Imam, Bisher. / Self-organizing linear output map (SOLO) : An artificial neural network suitable for hydrologic modeling and analysis. In: Water Resources Research. 2002 ; Vol. 38, No. 12. pp. 381-3817.
@article{a0a4d985982942ecb4d14bd3829a5a91,
title = "Self-organizing linear output map (SOLO): An artificial neural network suitable for hydrologic modeling and analysis",
abstract = "Artificial neural networks (ANNs) can be useful in the prediction of hydrologic variables, such as streamflow, particularly when the underlying processes have complex nonlinear interrelationships. However, conventional ANN structures suffer from network training issues that significantly limit their widespread application. This paper presents a multivariate ANN procedure entitled self-organizing linear output map (SOLO), whose structure has been designed for rapid, precise, and inexpensive estimation of network structure/parameters and system outputs. More important, SOLO provides features that facilitate insight into the underlying processes, thereby extending its usefulness beyond forecast applications as a tool for scientific investigations. These characteristics are demonstrated using a classic rainfall-runoff forecasting problem. Various aspects of model performance are evaluated in comparison with other commonly used modeling approaches, including multilayer feedforward ANNs, linear time series modeling, and conceptual rainfall-runoff modeling.",
keywords = "Artificial neural network, Overfitting, Principal component analysis, Rainfall-runoff modeling, Self-organizing feature map, SOLO",
author = "Hsu, {Kuo Lin} and Gupta, {Hoshin Vijai} and Xiaogang Gao and Soroosh Sorooshian and Bisher Imam",
year = "2002",
month = "12",
day = "1",
language = "English (US)",
volume = "38",
pages = "381--3817",
journal = "Water Resources Research",
issn = "0043-1397",
publisher = "American Geophysical Union",
number = "12",

}

TY - JOUR

T1 - Self-organizing linear output map (SOLO)

T2 - An artificial neural network suitable for hydrologic modeling and analysis

AU - Hsu, Kuo Lin

AU - Gupta, Hoshin Vijai

AU - Gao, Xiaogang

AU - Sorooshian, Soroosh

AU - Imam, Bisher

PY - 2002/12/1

Y1 - 2002/12/1

N2 - Artificial neural networks (ANNs) can be useful in the prediction of hydrologic variables, such as streamflow, particularly when the underlying processes have complex nonlinear interrelationships. However, conventional ANN structures suffer from network training issues that significantly limit their widespread application. This paper presents a multivariate ANN procedure entitled self-organizing linear output map (SOLO), whose structure has been designed for rapid, precise, and inexpensive estimation of network structure/parameters and system outputs. More important, SOLO provides features that facilitate insight into the underlying processes, thereby extending its usefulness beyond forecast applications as a tool for scientific investigations. These characteristics are demonstrated using a classic rainfall-runoff forecasting problem. Various aspects of model performance are evaluated in comparison with other commonly used modeling approaches, including multilayer feedforward ANNs, linear time series modeling, and conceptual rainfall-runoff modeling.

AB - Artificial neural networks (ANNs) can be useful in the prediction of hydrologic variables, such as streamflow, particularly when the underlying processes have complex nonlinear interrelationships. However, conventional ANN structures suffer from network training issues that significantly limit their widespread application. This paper presents a multivariate ANN procedure entitled self-organizing linear output map (SOLO), whose structure has been designed for rapid, precise, and inexpensive estimation of network structure/parameters and system outputs. More important, SOLO provides features that facilitate insight into the underlying processes, thereby extending its usefulness beyond forecast applications as a tool for scientific investigations. These characteristics are demonstrated using a classic rainfall-runoff forecasting problem. Various aspects of model performance are evaluated in comparison with other commonly used modeling approaches, including multilayer feedforward ANNs, linear time series modeling, and conceptual rainfall-runoff modeling.

KW - Artificial neural network

KW - Overfitting

KW - Principal component analysis

KW - Rainfall-runoff modeling

KW - Self-organizing feature map

KW - SOLO

UR - http://www.scopus.com/inward/record.url?scp=0036998831&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036998831&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0036998831

VL - 38

SP - 381

EP - 3817

JO - Water Resources Research

JF - Water Resources Research

SN - 0043-1397

IS - 12

ER -