Claim Missing Document
Check
Articles

Found 3 Documents
Search

Implementation of the Backpropagation Method to Predict the Percentage of Women as Professionals on the Island of Sumatra Tata Rizky Amalia; Solikhun Solikhun
International Journal of Mechanical Computational and Manufacturing Research Vol. 11 No. 2 (2022): August: Mechanical Computational And Manufacturing Research
Publisher : Trigin Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (794.922 KB) | DOI: 10.35335/computational.v11i2.5

Abstract

This study aims to obtain information on the best algorithm from the two algorithms that will be compared based on the smallest/lowest performance value or MSE value, which can later be used as a reference and information for solving women's problems as professional workers on the island of Sumatra. The data used in this study are women as professional workers (percent) 2012-2021 at the Central Statistics Agency (BPS). The algorithm used is Backpropagation Neural Network. Data analysis was carried out using the Artificial Neural Network method using Matlab R2011b(7.13) software. In this review, 5 structural models were used, namely: 4-10-1, 4-15-1, 4-20-1, 4-25-1, 4-30-1, out of five models.
Determining the Best Performance Using the Backpropagation Algorithm for Expenditure per Capita in North Sumatra Yogi Pratama; Solikhun Solikhun
International Journal of Mechanical Computational and Manufacturing Research Vol. 11 No. 2 (2022): August: Mechanical Computational And Manufacturing Research
Publisher : Trigin Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (561.067 KB) | DOI: 10.35335/computational.v11i2.6

Abstract

In an effort to maintain per capita income in Indonesia, the Government must take action through strengthening national protection. Per capita is the average income of all residents in a country. Per capita income is obtained from the distribution of the national income of a country by the total population of that country. There is a decrease in the population per capita of North Sumatra at the Central Statistics Agency (BPS) in 2020. The author will use the backpropagation algorithm to make a performance. Backpropagation iskone ofkmethodkartificial neural networklquite reliablejinlsolvekproblem. In researchj5 models are usedlarchitecture: 4-15-1, 4-30-1,k4-45-1, 4-60-1, 4.-75-1, fromjfive modelslThus, the architectural model 4 -75-1 provides the best accuracy withK452 iteration epochs and MSE is 0.00001536
Mushroom Production Prediction Model using Conjugate Gradient Algorithm Yosua Chandra Simamora; Solikhun Solikhun; Lise Pujiastuti; Mochamad Wahyudi
International Journal of Mechanical Computational and Manufacturing Research Vol. 11 No. 2 (2022): August: Mechanical Computational And Manufacturing Research
Publisher : Trigin Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (449.464 KB) | DOI: 10.35335/computational.v11i2.7

Abstract

Mushrooms are heterotrophic living things that act as saprophytes on dead plants. Mushrooms contain many important substances such as protein, amino acids, lysine, histidine, etc. Mushrooms tend to be better consumed than animal meat, even the content of lysine and histidine contained in mushrooms is greater than eggs. In recent years the volume of Mushroom Demand has increased, while production has decreased, especially on the island of Sumatra, namely in 2020 and 2021. Therefore, it is necessary to predict the estimated production of mushroom plants on the island of Sumatra so that the government on the island of Sumatra has clear data references to determine policies and make the right steps so that the production of mushroom plants on the island of Sumatra does not continue to decline. The method used in predicting is one of the ANN methods, namely the Conjugate Gradient Algorithm. The data used in this paper is Vegetable Crop Production data from 2014-2021 which was obtained from the website of the Central Statistics Agency. Based on this data, network architecture models such as 3-10-1, 3-15-1, 3-20-1, 3-25-1, 3-30-1, will be formed and defined. From the five models, training and testing values were obtained which showed that the most optimal architectural model was 3-10-1 with a Performance/MSE test value of 0.00055034. This value is the smallest of the 5 architectural models after the training and testing process. From this it can be concluded that this model can be applied to predict mushroom production on the island of Sumatra