Biography commercial serial crime analysis using enhanced dynamic neural network

Anahita Ghazvini, Mohd Zakree Ahmad Nazri, Siti Norul Huda Sheikh Abdullah, Md Nawawi Junoh, Zainal Abidin Bin Kasim

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

In sphere of criminology, suspect prediction analysis has been the point of convergence for many researchers. The focus of this study is on three prime attributes of next serial suspect's biography including nationality, age and time. Generally, to prevent the uncertainty in dynamic systems by nonlinear methods, a predictor is required in Time Delay Neural Network (TDNN). However, existing TDNN with single activation function is less effective to predict labeled class due to lower accuracy. Poor approximation of smooth mapping in single hidden layer makes it less effective. This study aims to propose a combined transfer functions to improve Nonlinear Autoregressive Time Series for performance prediction with exogenous (external) input (NARX)'s by utilizing Levenberg-Marquardt (LM) and Scaled Conjugate Gradient (SCG) algorithms. Consequently Hyperbolic Tangent Sigmoid (Tansig) and Radial Basis Function (RBF) are used in LM and SCG algorithms as bi-transfer functions for prediction of next suspect's biography in commercial serial case. The results of NARX model with combination of Tansig and RBF as two objective of transfer functions of LM and SCG, presented better performance for prediction of next serial crime suspect's biography in comparison to single activation function of Tansig and RBF.

Original languageEnglish
Title of host publicationProceedings of the 2015 7th International Conference of Soft Computing and Pattern Recognition, SoCPaR 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages334-339
Number of pages6
ISBN (Electronic)9781467393607
DOIs
Publication statusPublished - 15 Jun 2016
Event7th International Conference of Soft Computing and Pattern Recognition, SoCPaR 2015 - Fukuoka, Japan
Duration: 13 Nov 201515 Nov 2015

Other

Other7th International Conference of Soft Computing and Pattern Recognition, SoCPaR 2015
CountryJapan
CityFukuoka
Period13/11/1515/11/15

Fingerprint

Dynamic Neural Networks
Crime
Levenberg-Marquardt
Radial Functions
Transfer Function
Basis Functions
Conjugate Gradient Algorithm
Activation Function
Neural networks
Tangent line
Prediction
Time Delay
Transfer functions
Neural Networks
Hyperbolic tangent
Autoregressive Time Series
Nonlinear Time Series
Conjugate Gradient
Performance Prediction
Time delay

Keywords

  • Criminology
  • Modelling
  • NARX
  • Neural Network
  • Objective Function

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Control and Optimization
  • Modelling and Simulation

Cite this

Ghazvini, A., Ahmad Nazri, M. Z., Sheikh Abdullah, S. N. H., Junoh, M. N., & Abidin Bin Kasim, Z. (2016). Biography commercial serial crime analysis using enhanced dynamic neural network. In Proceedings of the 2015 7th International Conference of Soft Computing and Pattern Recognition, SoCPaR 2015 (pp. 334-339). [7492769] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SOCPAR.2015.7492769

Biography commercial serial crime analysis using enhanced dynamic neural network. / Ghazvini, Anahita; Ahmad Nazri, Mohd Zakree; Sheikh Abdullah, Siti Norul Huda; Junoh, Md Nawawi; Abidin Bin Kasim, Zainal.

Proceedings of the 2015 7th International Conference of Soft Computing and Pattern Recognition, SoCPaR 2015. Institute of Electrical and Electronics Engineers Inc., 2016. p. 334-339 7492769.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ghazvini, A, Ahmad Nazri, MZ, Sheikh Abdullah, SNH, Junoh, MN & Abidin Bin Kasim, Z 2016, Biography commercial serial crime analysis using enhanced dynamic neural network. in Proceedings of the 2015 7th International Conference of Soft Computing and Pattern Recognition, SoCPaR 2015., 7492769, Institute of Electrical and Electronics Engineers Inc., pp. 334-339, 7th International Conference of Soft Computing and Pattern Recognition, SoCPaR 2015, Fukuoka, Japan, 13/11/15. https://doi.org/10.1109/SOCPAR.2015.7492769
Ghazvini A, Ahmad Nazri MZ, Sheikh Abdullah SNH, Junoh MN, Abidin Bin Kasim Z. Biography commercial serial crime analysis using enhanced dynamic neural network. In Proceedings of the 2015 7th International Conference of Soft Computing and Pattern Recognition, SoCPaR 2015. Institute of Electrical and Electronics Engineers Inc. 2016. p. 334-339. 7492769 https://doi.org/10.1109/SOCPAR.2015.7492769
Ghazvini, Anahita ; Ahmad Nazri, Mohd Zakree ; Sheikh Abdullah, Siti Norul Huda ; Junoh, Md Nawawi ; Abidin Bin Kasim, Zainal. / Biography commercial serial crime analysis using enhanced dynamic neural network. Proceedings of the 2015 7th International Conference of Soft Computing and Pattern Recognition, SoCPaR 2015. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 334-339
@inproceedings{e2255bd038664a088825400f456545ca,
title = "Biography commercial serial crime analysis using enhanced dynamic neural network",
abstract = "In sphere of criminology, suspect prediction analysis has been the point of convergence for many researchers. The focus of this study is on three prime attributes of next serial suspect's biography including nationality, age and time. Generally, to prevent the uncertainty in dynamic systems by nonlinear methods, a predictor is required in Time Delay Neural Network (TDNN). However, existing TDNN with single activation function is less effective to predict labeled class due to lower accuracy. Poor approximation of smooth mapping in single hidden layer makes it less effective. This study aims to propose a combined transfer functions to improve Nonlinear Autoregressive Time Series for performance prediction with exogenous (external) input (NARX)'s by utilizing Levenberg-Marquardt (LM) and Scaled Conjugate Gradient (SCG) algorithms. Consequently Hyperbolic Tangent Sigmoid (Tansig) and Radial Basis Function (RBF) are used in LM and SCG algorithms as bi-transfer functions for prediction of next suspect's biography in commercial serial case. The results of NARX model with combination of Tansig and RBF as two objective of transfer functions of LM and SCG, presented better performance for prediction of next serial crime suspect's biography in comparison to single activation function of Tansig and RBF.",
keywords = "Criminology, Modelling, NARX, Neural Network, Objective Function",
author = "Anahita Ghazvini and {Ahmad Nazri}, {Mohd Zakree} and {Sheikh Abdullah}, {Siti Norul Huda} and Junoh, {Md Nawawi} and {Abidin Bin Kasim}, Zainal",
year = "2016",
month = "6",
day = "15",
doi = "10.1109/SOCPAR.2015.7492769",
language = "English",
pages = "334--339",
booktitle = "Proceedings of the 2015 7th International Conference of Soft Computing and Pattern Recognition, SoCPaR 2015",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Biography commercial serial crime analysis using enhanced dynamic neural network

AU - Ghazvini, Anahita

AU - Ahmad Nazri, Mohd Zakree

AU - Sheikh Abdullah, Siti Norul Huda

AU - Junoh, Md Nawawi

AU - Abidin Bin Kasim, Zainal

PY - 2016/6/15

Y1 - 2016/6/15

N2 - In sphere of criminology, suspect prediction analysis has been the point of convergence for many researchers. The focus of this study is on three prime attributes of next serial suspect's biography including nationality, age and time. Generally, to prevent the uncertainty in dynamic systems by nonlinear methods, a predictor is required in Time Delay Neural Network (TDNN). However, existing TDNN with single activation function is less effective to predict labeled class due to lower accuracy. Poor approximation of smooth mapping in single hidden layer makes it less effective. This study aims to propose a combined transfer functions to improve Nonlinear Autoregressive Time Series for performance prediction with exogenous (external) input (NARX)'s by utilizing Levenberg-Marquardt (LM) and Scaled Conjugate Gradient (SCG) algorithms. Consequently Hyperbolic Tangent Sigmoid (Tansig) and Radial Basis Function (RBF) are used in LM and SCG algorithms as bi-transfer functions for prediction of next suspect's biography in commercial serial case. The results of NARX model with combination of Tansig and RBF as two objective of transfer functions of LM and SCG, presented better performance for prediction of next serial crime suspect's biography in comparison to single activation function of Tansig and RBF.

AB - In sphere of criminology, suspect prediction analysis has been the point of convergence for many researchers. The focus of this study is on three prime attributes of next serial suspect's biography including nationality, age and time. Generally, to prevent the uncertainty in dynamic systems by nonlinear methods, a predictor is required in Time Delay Neural Network (TDNN). However, existing TDNN with single activation function is less effective to predict labeled class due to lower accuracy. Poor approximation of smooth mapping in single hidden layer makes it less effective. This study aims to propose a combined transfer functions to improve Nonlinear Autoregressive Time Series for performance prediction with exogenous (external) input (NARX)'s by utilizing Levenberg-Marquardt (LM) and Scaled Conjugate Gradient (SCG) algorithms. Consequently Hyperbolic Tangent Sigmoid (Tansig) and Radial Basis Function (RBF) are used in LM and SCG algorithms as bi-transfer functions for prediction of next suspect's biography in commercial serial case. The results of NARX model with combination of Tansig and RBF as two objective of transfer functions of LM and SCG, presented better performance for prediction of next serial crime suspect's biography in comparison to single activation function of Tansig and RBF.

KW - Criminology

KW - Modelling

KW - NARX

KW - Neural Network

KW - Objective Function

UR - http://www.scopus.com/inward/record.url?scp=84979246885&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84979246885&partnerID=8YFLogxK

U2 - 10.1109/SOCPAR.2015.7492769

DO - 10.1109/SOCPAR.2015.7492769

M3 - Conference contribution

AN - SCOPUS:84979246885

SP - 334

EP - 339

BT - Proceedings of the 2015 7th International Conference of Soft Computing and Pattern Recognition, SoCPaR 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -