TY - GEN
T1 - Predicting Vehicle Theft with Backpropagation Algorithm in East Java Regional Police
AU - Setyan, Ardath Prahara
AU - Affandi, Achmad
AU - Sumpeno, Surya
AU - Romahadi, Dedik
N1 - Publisher Copyright:
© 2021 IEEE.
PY - 2021/6/29
Y1 - 2021/6/29
N2 - Vehicle theft in East Java is a crime that causes unrest and anxiety in the public. The increased rate of vehicle theft is due to several factors, including economic, environment, low education, increased unemployment, and lacking legal awareness. One way to anticipate an increased level of theft is by predicting the possibility of theft based on previous incident data. This research is focused on the design of an intelligent system to predict the locations prone to vehicle theft using Backpropagation Neural Network. The training dataset used was obtained from data on the location of vehicle thefts in the East Java Regional Police from 2015 to 2019. Backpropagation architecture model 7-10-6 was used. The best level of accuracy for the performance of this model is 100%, with an epoch of 65 iterations, cross entropy of 5.188, and an error value of 0. Prediction test was carried out on ten new theft data in 2020 and 2021. The test resulted in a quite accurate level of accuracy in predicting locations prone to theft in a short time and gave correct prediction results.
AB - Vehicle theft in East Java is a crime that causes unrest and anxiety in the public. The increased rate of vehicle theft is due to several factors, including economic, environment, low education, increased unemployment, and lacking legal awareness. One way to anticipate an increased level of theft is by predicting the possibility of theft based on previous incident data. This research is focused on the design of an intelligent system to predict the locations prone to vehicle theft using Backpropagation Neural Network. The training dataset used was obtained from data on the location of vehicle thefts in the East Java Regional Police from 2015 to 2019. Backpropagation architecture model 7-10-6 was used. The best level of accuracy for the performance of this model is 100%, with an epoch of 65 iterations, cross entropy of 5.188, and an error value of 0. Prediction test was carried out on ten new theft data in 2020 and 2021. The test resulted in a quite accurate level of accuracy in predicting locations prone to theft in a short time and gave correct prediction results.
KW - artificial neural network
KW - backpropagation
KW - intelligent system
KW - prediction
KW - vehicle theft
UR - http://www.scopus.com/inward/record.url?scp=85112720197&partnerID=8YFLogxK
U2 - 10.1109/ICAICST53116.2021.9497819
DO - 10.1109/ICAICST53116.2021.9497819
M3 - Conference contribution
AN - SCOPUS:85112720197
T3 - ICAICST 2021 - 2021 International Conference on Artificial Intelligence and Computer Science Technology
SP - 19
EP - 24
BT - ICAICST 2021 - 2021 International Conference on Artificial Intelligence and Computer Science Technology
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2021 International Conference on Artificial Intelligence and Computer Science Technology, ICAICST 2021
Y2 - 29 June 2021
ER -