TY - GEN
T1 - Accuracy Rate of Relevance Vector Machine with Modified Algorithm
T2 - 4th International Conference on Engineering and Technology for Sustainable Development, ICET4SD 2021
AU - Syaharuddin,
AU - Fatmawati,
AU - Suprajitno, Herry
N1 - Publisher Copyright:
© 2023 American Institute of Physics Inc.. All rights reserved.
PY - 2023/12/29
Y1 - 2023/12/29
N2 - This research aims to analyze the accuracy of the Relevance Vector Machine (RVM) method, including algorithm modification and unmodified algorithms or other methods in the field of prediction. The data is collected from indexing databases such as Scopus, Sciencedirect, and Google Scholar. The criteria set are (1) articles published in 2010-2021, (2) search keywords "prediction, forecasting, relevance vector machine, RVM"; (3) existence coefficient correlation (R) value, accuracy rate, and the amount of data predicted (N). In addition, the data is analyzed using JASP software based on effect size (ES) and summary effect (SE) values. The data analysis showed that for the unmodified RVM method case, the accuracy rate of up to 30 data that meets the standard is 87% (range 0.80-0.93), using the Random Effects (RE) model, and if the RVM is modified, then the average accuracy rate is 93% (range 0.93-0.96). Finally, modifying the algorithm of the RVM method will have a great impact on the accuracy level in the prediction process.
AB - This research aims to analyze the accuracy of the Relevance Vector Machine (RVM) method, including algorithm modification and unmodified algorithms or other methods in the field of prediction. The data is collected from indexing databases such as Scopus, Sciencedirect, and Google Scholar. The criteria set are (1) articles published in 2010-2021, (2) search keywords "prediction, forecasting, relevance vector machine, RVM"; (3) existence coefficient correlation (R) value, accuracy rate, and the amount of data predicted (N). In addition, the data is analyzed using JASP software based on effect size (ES) and summary effect (SE) values. The data analysis showed that for the unmodified RVM method case, the accuracy rate of up to 30 data that meets the standard is 87% (range 0.80-0.93), using the Random Effects (RE) model, and if the RVM is modified, then the average accuracy rate is 93% (range 0.93-0.96). Finally, modifying the algorithm of the RVM method will have a great impact on the accuracy level in the prediction process.
UR - http://www.scopus.com/inward/record.url?scp=85182380878&partnerID=8YFLogxK
U2 - 10.1063/5.0164268
DO - 10.1063/5.0164268
M3 - Conference contribution
AN - SCOPUS:85182380878
T3 - AIP Conference Proceedings
BT - AIP Conference Proceedings
A2 - Dharma, Irfan Aditya
A2 - Puspasari, Ifa
A2 - Murnani, Suatmi
A2 - Sugarindra, Muchamad
A2 - Rahma, Fadilla Noor
PB - American Institute of Physics Inc.
Y2 - 13 December 2021
ER -