JOMLAI: Journal of Machine Learning and Artificial Intelligence
Focus and Scope JOMLAI: Journal of Machine Learning and Artificial Intelligence is a scientific journal related to machine learning and artificial intelligence that contains scientific writings on pure research and applied research in the field of machine learning and artificial intelligence as well as an overview of the development of theories, methods, and related applied sciences. Topics cover the following areas (but are not limited to): Software engineering Hardware Engineering Information Security System Engineering Expert system Decision Support System Data Mining Artificial Intelligence System Computer network Computer Engineering Image processing Genetic Algorithm Information Systems Business Intelligence and Knowledge Management Database System Big Data Internet of Things Enterprise Computing Machine Learning Other relevant study topics Noted: Articles have primary citations and have never been published online or printed before
Articles
10 Documents
Search results for
, issue
"Vol. 1 No. 3 (2022): September"
:
10 Documents
clear
<![CDATA[Application of Artificial Neural Networks to Predict Exports of Four-Wheeled Vehicles by Destination Country ]]>
<![CDATA[Ema Meyliza]]>;
<![CDATA[Eka Irawan]]>;
<![CDATA[Dedi Suhendro]]>
<![CDATA[ JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 1 No. 3 (2022): September ]]>
Publisher : <![CDATA[Yayasan Literasi Sains Indonesia ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (450.585 KB)
|
DOI: 10.55123/jomlai.v1i3.914
<![CDATA[Motorized vehicles are vehicles that are energized through machines and used for land transportation,, as well as movement through technical equipment in the form of electric motors and other tools that have the function of converting an energy source into power to drive the vehicle. This study aims to determine the results of the predicted number of four-wheeled vehicle exports by destination country in the years to come. The research data used is data on exports of four-wheeled motor vehicles by destination country for 2012-2020. The algorithm used in this research is an artificial neural network with Back-propagation method. The best architectural model used in this research is architect 4-4-1 with an accuracy rate of 82% epoch of 2261 iterations and MSE of 0.0081876. So it can be concluded that the model can be used to predict export data for four-wheeled vehicles by destination country.]]>
<![CDATA[Implementation of the Weighted Moving Average Method for Forecasting the Production of Manila Duck meat in Indonesia ]]>
<![CDATA[Diana Pratiwi]]>;
<![CDATA[Riki Winanjaya]]>;
<![CDATA[Irawan Irawan]]>
<![CDATA[ JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 1 No. 3 (2022): September ]]>
Publisher : <![CDATA[Yayasan Literasi Sains Indonesia ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (723.134 KB)
|
DOI: 10.55123/jomlai.v1i3.916
<![CDATA[Manila duck is a waterfowl originating from South America, through the Philippines this type of duck entered Indonesia and has a large distribution in various regions of Indonesia the production on manila duck meat and from 2019-2020 has decreased due to the covid-19 pandemic which resulted in economic difficulties. And the lack of demand from restaurants and households so that the amount of production decrease. However, in 2020-2021 production will increase due to the relaxation from the previous pandemic and the demand and marketing has increased to that the number of production has increased from the previous year. The Weighted Moving Average method is a method used to determine the latest trend with a moving average value. The purpose of this study was to analyses the amount of production of manila duck meat in solving the problem. The result obtained with the smallest error percentage are at F128 in the province of North Maluku with MAPE value of 0,003 or equal to 0,3% with a bias of -0,25, MAD 0,25, MSE 0,06, with a forecasting value of 83,29 which is close to the original data, namely 83,04 so that the forecast value for 2022 is 83,24 tons.]]>
<![CDATA[Forecasting of Rubber Production in North Sumatra with Backpropagation Algorithm ]]>
<![CDATA[Josua Fernando Simanjuntak]]>;
<![CDATA[Riki Winanjaya]]>;
<![CDATA[Wendi Robiansyah]]>
<![CDATA[ JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 1 No. 3 (2022): September ]]>
Publisher : <![CDATA[Yayasan Literasi Sains Indonesia ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (693.975 KB)
|
DOI: 10.55123/jomlai.v1i3.917
<![CDATA[Rubber is a commodity to produce tires, balloons, and other rubber-based products. Indonesia is the second largest rubber producer and distributor in the world. But, rubber production level tends to fluctuate. Therefore, an analysis is needed to predict rubber production in the future thus rubber plantations, especially folk-owned, can take steps to prevent if declines in production are found. One way that can be done to predict is by utilizing Artificial Neural Network with Backpropagation method, since it provides accurate results. In this research, 10 network architecture models were tested and the best architecture achieved was 10-10-11-1 with accuracy of 96%. With that architecture, predictions are done and resulted in estimated rubber production in North Sumatra for 2021-2025.]]>
<![CDATA[Artificial Neural Network Method in Predicting the Amount of Manila Duck Meat Production by Province in Indonesia ]]>
<![CDATA[Joko Pamungkas]]>;
<![CDATA[Riki Winanjaya]]>;
<![CDATA[Wendi Robiansyah]]>
<![CDATA[ JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 1 No. 3 (2022): September ]]>
Publisher : <![CDATA[Yayasan Literasi Sains Indonesia ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (646.291 KB)
|
DOI: 10.55123/jomlai.v1i3.918
<![CDATA[Duck meat is a source of animal protein that many Indonesians need because it can increase nutritional needs to improve people's quality of life. One of the types of ducks used in this study is the Manila duck, this type of duck was chosen because it is very easy to maintain and the price is also relatively affordable. Based on data on the production of Manila ducks in Indonesia from several provinces, the annual production amount is unstable. Therefore, it is important to make predictions about this matter as information for the government. The data sample used in this study is manila duck production data taken from the Indonesian Central Statistics Agency in 2017-2020. This research uses backpropagation algorithm. Based on the results of the analysis, the best architectural model is 3-6-1 which will later be used to predict the amount of manila duck meat production in 2022 because it has the highest accuracy rate compared to other models, which is 74%. MSE Testing is 0,00412. Based on this model, predictions of the amount of manila duck meat production will be made based on provinces in Indonesia. From the prediction results, it can be seen that there are 25 provinces that are estimated to experience an increase in production in 2022 or around 73,5% (25 provinces) of a total of 34 provinces in Indonesia. Meanwhile, 9 other provinces experienced a decline or around 26,5%.]]>
<![CDATA[Application of Artificial Neural Networks in Predicting Salt Imports by Country of Origin Using the Back-propagation Method ]]>
<![CDATA[Sari Marito Tondang]]>;
<![CDATA[Heru Satria Tambunan]]>;
<![CDATA[Susiani Susiani]]>
<![CDATA[ JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 1 No. 3 (2022): September ]]>
Publisher : <![CDATA[Yayasan Literasi Sains Indonesia ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (401.659 KB)
|
DOI: 10.55123/jomlai.v1i3.919
<![CDATA[Salt is a basic consumption material needed by the community and various industries. Indonesia is a country that has many beaches that have great potential as a source of salt production. But Indonesia is still dependent on imports so that industrial imports continue to increase, can directly or indirectly affect the risk of the country’s economic pattern. An increase in salt imports although there was also a decrease but only slightly and did not last long from several countries from 2010-2020 recorded in the Central Statistics Agency (BPS). In this study, the author will predict the import of salt for the next 3 years using the Back-propagation algorithm. Back-propagation is one of the artificial neural network methods that is quite reliable in solving problems where the network tries to achieve stability again to achieve the expected output and there is a learning process by adjusting connection weights. This study uses 6 architectural models : 5-80-1, 5-90-1, 5-100-1, 5-110-1, from the four models the best architectural model is obtained namely 5-90-1 with an accuracy value of 75%, epoch 4265 iterations, and MSE Testing 0,01569.]]>
<![CDATA[Implementation of One-step Secant Algorithm for Forecasting Open Unemployment by Highest Educational Graduate ]]>
<![CDATA[Ismi Azhami]]>;
<![CDATA[Eka Irawan]]>;
<![CDATA[Dedi Suhendro]]>
<![CDATA[ JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 1 No. 3 (2022): September ]]>
Publisher : <![CDATA[Yayasan Literasi Sains Indonesia ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (601.346 KB)
|
DOI: 10.55123/jomlai.v1i3.946
<![CDATA[Based on data, the open unemployment rate according to the highest education graduate in Indonesia shows the number of semester unemployment which has an unstable value, sometimes up and sometimes down. This study aims to implement the ability and performance of one of the training functions on the backpropagation algorithm, namely one-step secant, which can later be used as a reference in terms of data forecasting. The one-step secant algorithm is an algorithm that is able to train any network as long as the input, weight and transfer functions have derivative functions and this algorithm is able to make training more efficient because it does not require a very long time. The data used in this study is open unemployment data according to the highest education completed in Indonesia in 2006-2021 based on semester, which is sourced from the Indonesian Central Statistics Agency. Based on this data, a network architecture model will be formed and determined using the One-step secant method, including 14-13-2, 14-16-2, 14-19-2, 14-55-2, and 14-77- 2. From these 5 models, after training and testing, the results show that the best architectural model is 14-19-2 (14 is the input layer, 19 is the number of neurons in the hidden layer and 2 is the output layer). The accuracy level of the architectural model for semester 1 and semester 2 is 75% with MSE values of 0.00130797 and 0.00388535.]]>
<![CDATA[Classification Techniques in Predicting New Student Admission Using the Naïve Bayes Method ]]>
<![CDATA[Suwayudhi Suwayudhi]]>;
<![CDATA[Eka Irawan]]>;
<![CDATA[Bahrudi Efendi Damanik]]>
<![CDATA[ JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 1 No. 3 (2022): September ]]>
Publisher : <![CDATA[Yayasan Literasi Sains Indonesia ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (394.074 KB)
|
DOI: 10.55123/jomlai.v1i3.963
<![CDATA[Admission of new students is the registration process for new students entering school and the initial gate through which students enter the object of Education; this activity is the starting point for determining the smoothness of the tasks of a school, assisted by teaching staff and equipped with optimal facilities and infrastructure in teaching and learning activities, producing skilled and broad-minded students. However, the uncertainty of the number of registrants also influences the policies that will be taken in the future. Therefore, it is necessary to forecast or predict to estimate the number of students who are likely to register so that the school can prepare everything. In this study, the prediction process for new students will use a classification technique using the Naïve Bayes method. This study aims to predict the rise and fall of the number of students who register using the Naïve Bayes method. The research data was obtained by distributing questionnaires randomly to 200 respondents (students) who were about to enter high school. The data is accumulated using the help of Microsoft excel. The results obtained are that the prediction of high-class precision is 100%, while the prediction of low-class precision is 94.23%. The conclusion is that the extracurricular, cost and distance criteria need attention and improvement. This is because disinterest and low prediction are higher than interest with high prediction results.]]>
<![CDATA[Classification of Internet Addiction Levels in Students Using the Naïve Bayes Algorithm ]]>
<![CDATA[Fakhriyah Zulfah Parinduri]]>;
<![CDATA[Rafika Dewi]]>;
<![CDATA[Susiani Susiani]]>
<![CDATA[ JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 1 No. 3 (2022): September ]]>
Publisher : <![CDATA[Yayasan Literasi Sains Indonesia ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (489.824 KB)
|
DOI: 10.55123/jomlai.v1i3.965
<![CDATA[The presence of the internet on students has a big influence on science and technology which makes the internet as an additional insight to find the information needed, apart from being a source of information, students also access the internet as a means of entertainment. So that it makes students last longer in front of gadgets or computers continuously. The purpose of this study is to determine whether students are indicated by internet addiction and provide input to STIKOM Tunas Bangsa to make policies that use the internet as a learning process so that internet addiction does not occur excessively. Because it is very influential in the learning process to add insight about science and technology to students. The subjects carried out by this study were students who were studying at STIKOM Tunas Bangsa. Therefore, the research was conducted using the Naïve Bayes algorithm classification, in which the data was obtained using a questionnaire distributed to students. The subjects carried out by this study were students who were studying at STIKOM Tunas Bangsa. Therefore, the research was conducted using the Naïve Bayes algorithm classification, in which the data was obtained using a questionnaire distributed to students. It is hoped that this research can be information for students to be able to maintain self-control in utilizing various entertainments on the internet.]]>
<![CDATA[Movidius Neural Compute Stick for Real Time Detection of Human Objects with the Mobilenet-SSD Method ]]>
<![CDATA[Maulia Rahman]]>;
<![CDATA[Dedi Leman]]>
<![CDATA[ JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 1 No. 3 (2022): September ]]>
Publisher : <![CDATA[Yayasan Literasi Sains Indonesia ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (534.913 KB)
|
DOI: 10.55123/jomlai.v1i3.1025
<![CDATA[The presence of surveillance cameras plays an important role in helping the process of monitoring and evaluating human activities in the monitored area. This ability can prevent or trace undesirable events such as criminal acts or some accidents that related to human activities. However, most of the surveillance camera that used nowadays only held a passive role in security that can lead to an increased potential risk of negligence by the guards (users) in the process of monitoring the activities that are happening. This study aims to design a system that is able to improve the performance of surveillance cameras in detecting and calculating numbers of human based on Movidius NCS on a Raspberry Pi device so that the camera can be active and be able to provide optimal results and reduce the use of excess space on the storage. The human object detection system that is used in this research applies Deep Learning technique with Mobilenet-SSD as its network architecture model. The research trials were carried out under various conditions of light intensity starting from 50-550 lux and distance to objects in range of 1-10 meters. The results showed that the accuracy obtained by the system was able to reach 91.67% with 49.24% of storage efficiency.]]>
<![CDATA[Comparison of Borda and NRF (Normalized Rating Frequency) in Recommender System ]]>
<![CDATA[Taufiq Abidin]]>;
<![CDATA[Slamet Wiyono]]>
<![CDATA[ JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 1 No. 3 (2022): September ]]>
Publisher : <![CDATA[Yayasan Literasi Sains Indonesia ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (225.707 KB)
|
DOI: 10.55123/jomlai.v1i3.1026
<![CDATA[The Collaborative Filtering method is a popular method in making recommender systems. Although CF is a popular method, it has major problems, namely cold start and sparsity . Several studies have been conducted to treat cold starts and sparsity. One way to overcome cold start and sparsity is the Borda calculation method. Research using the Borda method has been carried out a lot but has not utilized the rating optimally. The NRF method is a new method offered to maximize the use of ratings. By using dummy test data, the NRF method is more effective than Borda in calculating recommendation scores.]]>
