IMPLEMENTATION OF DATA MINING WITH CLASSIFICATION AND FORECASTING METHOD USE MODEL GAUSSIAN NAÏVE BAYES FOR BUILDING STORE (STUDI CASE: TB SINAR JAYA)

Abstract

In the construction industry, the building materials business is a necessity, one of the businesses that is currently growing rapidly is the property business. The public’s need for a place to live becomes a business opportunity sought by the public. The Industry 4.0 transformation brought building materials stores to the online market, one of which is Tokopedia. TB. Sinar Jaya has a dataset of inventory and sales with a total of approximately 15,000 data rows which are updated each month. With large amounts of data, data mining and machine learning methods are needed in data management. 5 years rapid development of TB. Sinar Jaya has not been without problems, such as competition with online stores that offer lower prices than offline stores and a lack of strong marketing strategy. In this case, TB. Sinar Jaya wants help in making marketing strategy decisions by utilizing information system technology and minimizing existing problems. Based on these problems, it is necessary to implement data mining and machine learning gaussian algorithms naïve bayes to find out the average prices available on Tokopedia to increase sales and carry out classification, forecasting and TSA (Time Series Analysis) at TB Sinar Jaya. Based on the results of testing/research, the gaussian naive Bayes algorithm has good accuracy results with an accuracy level of 0.71 and gains insight, that is, for potential buyers they can do Wishlist efficiency, and for segments that generate profits below 50% of the total profit the researcher recommends carrying out a campaign program according to customer profile in order to improve the resulting profitability.