IMPLEMENTATION OF DATA MINING TO PREDICT THE VALUE OF INDONESIAN OIL AND NON-OIL AND GAS IMPORT EXPORTS USING THE LINEAR REGRESSION METHOD

Abstract

Indonesia's export-import activities in recent years, the value of Indonesia's exports and imports has decreased due to global conditions. The problems that occur are the uncertainty and complexity in estimating the value of international trade in the oil and gas and non-oil and gas sectors, dependence on just one or a few markets, and the problem of unfair competition, unfair competition between business actors can reduce export-import prices. The value of oil and gas and non-oil and gas exports and imports is influenced by several external factors that are difficult to predict, such as fluctuations in oil and gas prices, changes in trade policies, and global economic factors. The prediction results are obtained every month from the export value data using the rapid miner application. From the export data, the value of non-oil and gas exports obtains a very high value compared to the export data of oil and gas values. Then the results from rapid miner using the linear regression algorithm are obtained. The predicted import value of oil and gas and non-oil and gas value data in June is 209,162,268, and the predicted export value of oil and gas and non-oil and gas value data in June is 349,285,781 and non-oil and gas which more are predicted to have the highest value compared to the value of oil and gas in each month.