DATA MINING TO PREDICT INVENTORY AT DAPUR PINTAR STORE USING MULTIPLE LINEAR REGRESSION METHODS

Abstract

Dapur pintar store is a shop that sells household furniture. However, there are several obstacles faced by Dapur Pintar Store, such as the shop still having difficulty predicting future inventory, and the absence of the application of data mining in predicting inventory. This is interesting to solve in order to make it easier for the store to determine inventory predictions at Dapur Pintar Store. Utilizing data mining, especially using linear regression, can provide valuable insight for predicting sales of household products at Smart Kitchen Stores. Identify and collect relevant data for sales analysis. This includes historical sales data, customer demographic data, promotions or discounts applied, and other factors that may influence household product sales. Perform data preprocessing to clean incomplete or inaccurate data. In addition, perform data transformations such as normalization to ensure consistency and accuracy in analysis. Apply a linear regression model to understand the linear relationship between the independent variable (for example, time, promotion, or demographics) and the dependent variable (household product sales). This model can be used to make predictions based on historical patterns. Use historical data to train a linear regression model. The training process involves adjusting the model parameters to fit the training data, so that it can provide accurate predictions. Validate the model using data that was not used in training to ensure that the model can provide good predictions on new data. This helps avoid overfitting and ensures generalization of the model. In making inventory predictions for goods, we will use the Multiple Linear Regression method. Software that will be used to support data processing is RapidMiner.