Selamat Datang Di Repository Universitas Royal

ABSTRACT

APPLICATION OF MACHINE LEARNING WITH LINEAR REGRESSION ALGORITHM TO PREDICT SALES OF BANTU TANI SHOP

By: Gitara Yuninda Wardani (21.22.0369)
In the era of digitalization, effective data management is key to business decision-making. One of the main challenges facing Toko Bantu Tani is the significant fluctuation in demand due to price changes and other external factors, leading to stock imbalances. To overcome this problem, this research applies the Linear Regression algorithm in Machine Learning to predict the number of sales based on historical data, with the aim of improving the efficiency of stock planning and reducing the risk of excess or shortage of goods. This research uses a quantitative approach with the Linear Regression method that models the relationship between the independent variable (product price) and the dependent variable (quantity sold). The system developed is web-based, using Flask as a backend framework and MySQL as a database to store historical data and prediction results. Sales data is taken from the period November 2023 to November 2024 and analyzed to produce an optimal prediction model. The test results show that the prediction model has a high level of accuracy, with a Mean Absolute Percentage Error (MAPE) value below 5%, which indicates that the prediction has a low error. Evaluation is also carried out using Mean Squared Error (MSE) and Root Mean Squared Error (RMSE) to ensure the accuracy of the model. The implementation of this system is expected to help Toko Bantu Tani in managing stock better and improving business efficiency through data-based decision making.
Keywords: linear regression; machine learning; stock management; sales prediction; shop bantu tani