Price Prediction System of Basic Commodities Using Long Short-Term Memory Method: Analysis and Implementation for Future Projections

: The uncontrollable prices of basic commodities, especially food commodities, have resulted in losses for producers and consumers in the city of Cirebon. To be able to bridge these problems, it is necessary to make the right decisions. A prediction system is one of the elements that can be used to support the right decision-making. With an accurate price prediction system for staple food commodities, it is hoped that decision-making will be able to decide on good policies from the Cooperative, Small and Medium Enterprises Service for Trade and Industry (DKUKMPP) for the people of Cirebon City. The method used is Long Short-Term Memory (LSTM). The data used for the basic prices of food commodities, namely the last 3 years in 2020 - 2022, are sourced from the official website of the Ministry of Trade's Market Monitoring System and Basic Needs using Google Colab tools. 3. Predictions for staple food prices over the last 3 years include rice experiencing an increase of Rp. 275 per percentage of 2.35%, Granulated Sugar experienced an increase of Rp. 1000 with a percentage of 7.7%, Cooking Oil experiencing an increase of Rp. 1200 percent of 9.2%, Flour experienced a difference in the increase of Rp. 3500, a percentage of 50%, Beef Rp. 30,000 a proportion of 26%, Chicken Rp. 0, a percentage of 0%, Chicken Eggs, Rp. 9,000, a percentage of 45%, Red Chilies, Rp. 1,0,000, a percentage of 25%, Shallots, Rp. - 2,700 percent - 9.9%, Garlic IDR 11000 percentage 36.5%.


INTRODUCTION
DKUKMPP explains that the needs for clothing, food and shelter are three important basic human needs.Of the three basic needs, there is one basic need that is very important, namely the need for food.Food needs themselves refer to human needs for adequate and

METHOD Data Collection
The data used in this study is secondary data in the form of time series data, namely data on 10 prices of basic food commodities in DKUKMPP Cirebon City starting from January 1, 2020 -July 31, 2023.Method: Analysis and Implementation for Future Projections

Software development methods
The software development method used in this study uses the waterfall method which is a linear sequential software development method or classic life cycle [4].This method has 5 (five) interrelated stages, can be seen in Figure 2.

Design
At this stage, an analysis and design will be carried out on the things needed in the system to be built.From the results of the analysis, a system design will be made using a flowchart diagram that will be used as a reference to develop a price prediction system for food commodities.

Implementation (Implementasi)
The next stage is the implementation stage, where at this stage is the process of translating the entire system design that has been designed into program codes to produce a system as a whole.

Verification
The next stage is the verification stage, at the stage of testing the price prediction system of food commodities at DKUKMPP by simulating the price prediction system of staples using the Long Short-Term Memory (LSTM) method.Then see whether the input data from the food commodity price prediction system can predict and distinguish between the price of staples or not.Furthermore, the data will be presented in the form of a test graph of the price of food commodity staples.

Maintenance
The last stage is the maintenance or maintenance stage of the system, in which there is an installation process and the process of repairing the system to predict the price of food commodities if errors are found that are not detected at the verification stage.After passing this last stage will result in a better system.

RESULTS AND DISCUSSION
This section explains the results of the design that has been made and the discussion of the data used in this study.

Staple
Staples are basic food needs for human survival.The existence of staples is influenced by production and selling prices.In each region, usually the need for main food ingredients also varies.

Food Commodities
Food commodities are superior trade goods to support the economy of a community in an agricultural country.

A country's export activities can be
Rifqi Fahrudin 1 Kusnadi 2 Chandra Lukita 3 | 1622 ISSN: 2088-589X dominated by commodities in the form of agricultural food.In a broad sense, food is another term for food.

Prediction System
A system is an order consisting of several functional components (with specific tasks or functions) that are

Test Results
The results obtained from the implementation of the LSTM algorithm in predicting the price of basic commodities will be explained in the section below.This testing process uses 1044 data records.After the dataset is determined, it will enter the next process, namely the pre-processing Method: Analysis and Implementation for Future Projections stage, if the dataset is good, then proceed to divide it into 80% training data and 20% testing data.From the results of experiments conducted on the dataset of 10 prices of food commodities with a total of 1305 commodities.Then the difference and price shown in the picture are as follows: 1.The results feature section will display an explanation of the price prediction results of 10 staples from January 1, 2020, to July 31, 2023.

The table features section explains
the names of the staples of 10 staples, the mean or average of the price, the minimum price, the maximum price, data from the first price to be predicted and finally price data, price differences, and price percentages that function to find out all the results of the predicted prices of staples.
3. And The benefits feature section explains that staples have benefits not just necessities.

Difference Display Page
balanced nutrition every day.The price of basic food commodities is one of the problems currently being faced, where this problem is caused by data on prices of basic food commodities which are uncertain and not presented well because they are in the form of estimates from sellers to consumers with prices that may vary[1].Prices of basic commodities related to food commodities that will be predicted in this research are 10 commodities including rice, granulated sugar, cooking oil, wheat flour, beef, chicken meat, chicken eggs, red chilies, shallots, and garlic.Of the 10 basic commodities, these basic commodities are an initial reference that the price of basic commodities is an important thing in people's lives in Cirebon City.Where the need for basic food commodities for the community will continue to increase every day, month, and year.So, the Cirebon City DKUKMPP needs a price prediction system for basic commodities in Cirebon City with the aim of deciding on the exact price of basic food commodities whether they are rising, falling or stable.In previous research entitled "Prediction of Food Commodity Prices Using the Long Short-Term Memory Algorithm" [2], the research explains that predictions of food commodity prices have been carried out and can help farmers in knowing fluctuating food prices, especially commodity prices in traditional markets such as Sweet Market and Wage Market.In other research, "Short Term National Staple Material Price Prediction Using ARIMA" with ARIMA results can be used to predict national staple food prices in the short term between 1 to 30 days in the future.Experiments show that the ARIMA model built is able to predict prices of basic commodities quite accurately with an average error of 2.22% [3].This research aims to produce a prediction system for 10 staple food commodities in Cirebon City with data used from January 1, 2020, to July 31, 2023.The resulting predictions can be used by the Cirebon City DKUKMPP for decision making in determining the prices of these 10 staple food commodities.

Figure 1 .
Figure 1.Graph Dataset 10 Prices of Food Commodity Staples

Rifqi Fahrudin 1
Figure 2. Waterfall Development Method Method: Analysis and Implementation for Future Projections In making the "Basic Commodity Price Prediction System Using the LSTM Method (Case Study: DKUKMPP Cirebon City)", the first process is to take the price of food commodities which then the results of taking the prices of food commodities will be entered in excel and convert csv files and will be analyzed by applying the Long Short Term Memory method (LSTM) to determine the price of food commodity staples in the form of food commodity charts or not.The results of these calculations will be used as parameters to determine the prediction of prices of basic commodities.Graphs and tables of basic food prices will rise or fall when food commodities are predicted using the Long Short-Term Memory (LSTM) method.
interconnected and together aim to fulfill a certain process / work.Prediction is something that is done systematically to predict something that may happen in the future based on past and present information, so that the difference between something that happens, and the expected outcome can be minimized.Predictions don't have to give a definitive answer to an upcoming event but try to find the answer as close to what will happen as possible.Long Short-Term Memory (LSTM) Long Short-Term Memory (LSTM) is a popular Deep Learning algorithm that is suitable for making predictions and classifications related to time.This algorithm can be said to be a development or one type of RNN (Recurrent Neural Network) algorithm.In the RNN algorithm, the output from the last step is fed back as input on the currently active step.However, the RNN algorithm has the disadvantage that it cannot predict words stored in longterm memory.The structure of the LSTM algorithm consists of a neural network and several different memory blocks.These memory blocks are referred to as cells.The state of the cell and the hidden state will be passed to the next cell.below,the blue rectangular shape is an illustration of the LSTM cell.

Figure 4 .Figure 5 .
Figure 4.The Price of Tree Materials