Enhancing Sales Prediction for MSMEs: A Comparative Analysis of Neural Network and Linear Regression Algorithms
The increasingly fierce competition in the Micro, Small, and Medium Enterprises (MSME) industry has made business actors predict sales to find out future sales predictions and prepare strategies to deal with market trends that will occur in the future. Most MSMEs still do not have a prediction syste...
Saved in:
| Published in: | Jurnal Teknologi dan Manajemen Informatika Vol. 10; no. 1; pp. 81 - 91 |
|---|---|
| Main Authors: | , |
| Format: | Journal Article |
| Language: | English Indonesian |
| Published: |
Universitas Merdeka Malang
25.06.2024
|
| Subjects: | |
| ISSN: | 1693-6604, 2580-8044 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The increasingly fierce competition in the Micro, Small, and Medium Enterprises (MSME) industry has made business actors predict sales to find out future sales predictions and prepare strategies to deal with market trends that will occur in the future. Most MSMEs still do not have a prediction system. So, to set sales targets each year, they always use manual estimates by reviewing the previous year's sales data. Therefore, this research aims to predict sales and analyze the error value of sales data forecasting so that it can provide recommendations for strategies to increase sales. This research will apply neural network and linear regression algorithms to predict sales from 2020 to 2022. Based on the results of method testing, the artificial neural network algorithm is more suitable for forecasting sales than the linear regression algorithm. The test results obtained an RMSE value of 40,070 in the neural network method using one hidden layer and an RMSE value of 66,998 derived from the feature selection T-test and iterative T-test with a minimum tolerance value of 0.05 in the linear regression method. |
|---|---|
| ISSN: | 1693-6604 2580-8044 |
| DOI: | 10.26905/jtmi.v10i1.11875 |