Andharluana, Fiqih Pavita (2025) PENERAPAN METODE RADIAL BASIS FUNCTION NEURAL NETWORK UNTUK MEMPREDIKSI INDEKS HARGA KONSUMEN DI KOTA SURABAYA DENGAN FUNGSI AKTIVASI GAUSSIAN DAN MULTIKUADRATIK. Undergraduate thesis, Universitas Pembangunan Nasional Veteran Jawa Timur.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Cover)
21083010042_Cover.pdf Download (1MB) |
|
Text (Bab I)
21083010042_BAB I.pdf Download (175kB) |
|
Text (Bab II)
21083010042_BAB II.pdf Restricted to Registered users only until 19 June 2028. Download (564kB) |
|
Text (Bab III)
21083010042_BAB III.pdf Restricted to Registered users only until 19 June 2028. Download (414kB) |
|
Text (Bab IV)
21083010042_BAB IV.pdf Restricted to Registered users only until 19 June 2028. Download (1MB) |
|
Text (Bab V)
21083010042_BAB V.pdf Restricted to Registered users only until 19 June 2028. Download (144kB) |
|
Text (Daftar Pustaka)
21083010042_DAFTAR PUSTAKA.pdf Download (151kB) |
|
Text (Lampiran)
21083010042_LAMPIRAN.pdf Restricted to Registered users only until 19 June 2028. Download (240kB) |
Abstract
The Consumer Price Index (CPI) is the main indicator in measuring the inflation rate of a region by monitoring the prices of goods and services consumed by households. Unmentioned changes in the CPI value have a direct impact on people's purchasing power. The city of Surabaya, as a region with the third highest Gross Regional Domestic Product (GRDP) in Indonesia, has a high economic growth rate, so it requires proper inflation control, one of which is through accurate CPI prediction. Because an unstable CPI will affect the inflation rate, and an unstable inflation rate can disrupt economic stability. This study aims to compare the performance of two activation functions in the Radial Basis Function Neural Network (RBFNN) model, namely Gaussian and Multiquadratic, in predicting the CPI in Surabaya. Because it is still limited to previous studies. The RBFNN model is built by determining clusters, spread values, activation functions, weights, and outputs, and evaluated using Symmetric Mean Absolute Percentage Error (sMAPE). Based on the analysis results, it was obtained that the Multiquadratic activation function provided the best prediction results with a SMAPE value of 0.70%, which indicates a very high level of accuracy. The predicted CPI results for January to May 2025 are 107.61, 108.09, 108.54, 108.95, and 109.32, respectively.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > Q Science (General) Q Science > QA Mathematics Q Science > QA Mathematics > QA76.6 Computer Programming |
||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Data Science | ||||||||||||
| Depositing User: | Fiqih Pavita Andharluana | ||||||||||||
| Date Deposited: | 20 Jun 2025 02:28 | ||||||||||||
| Last Modified: | 20 Jun 2025 02:28 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/38607 |
Actions (login required)
 |
View Item |