Brahmana : Jurnal Penerapan Kecerdasan Buatan
BRAHMANA: Jurnal Penerapan Kecerdasan Buatan adalah sebuah jurnal peer-review secara online yang diterbitkan bertujuan sebagai sebuah forum penerbitan tingkat nasional di Indonesia bagi para peneliti, profesional, Mahasiswa dan praktisi dari berbagai bidang Ilmu Kecerdasan Buatan. BRAHMANA: Jurnal Penerapan Kecerdasan Buatan menerbitkan hasil karya asli dari penelitian terunggul dan termaju pada semua topik yang berkaitan dengan Kecerdasan Buatan. BRAHMANA: Jurnal Penerapan Kecerdasan Buatan terbit 2 (dua) nomor dalam setahun. Artikel yang telah dinyatakan diterima akan diterbitkan dalam nomor In-Press sebelum nomor regular terbit. BRAHMANA: Jurnal Penerapan Kecerdasan Buatan telah terindeks Google Scholar dan terus akan diupdate mengikuti perkembangan.
Articles
133 Documents
<![CDATA[Memprediksi Tingkat Penjualan Smartphone Apple di Indonesia Dengan Menggunakan Metode Backpropagation ]]>
<![CDATA[Abet Nego Situmorang]]>;
<![CDATA[Fathur Dwi Putra]]>;
<![CDATA[Jumanto Geogan Simanjuntak]]>;
<![CDATA[P Poningsih]]>
<![CDATA[ Brahmana : Jurnal Penerapan Kecerdasan Buatan Vol 4, No 1 (2022): Edisi Desember ]]>
Publisher : <![CDATA[LPPM STIKOM Tunas Bangsa ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.30645/brahmana.v4i1.107
<![CDATA[A system to predict the sales level of Apple branded smartphones in Indonesia. Artificial Neural Network is a method capable of performing mathematical processes to predict the sales level of Apple branded smartphones in Indonesia. Using the backpropagation method, the previous data processing is carried out to be used as input to predict the sales level of Apple smartphones in Indonesia. The data processed as variables are Apple Vendor Market. The data was taken from April 2021 to March 2022. April 2021 to September 2021 is used as input data, while October 2021 to March 2022 is used as target data. Several steps of Backpropagation are by initializing data, activation, calculating input data and output bias and data and bias changes. Those stages will obtain an output that will be achieved by having the smallest error value so that the prediction of the sales level of Apple branded smartphones in Indonesia is obtained. The training and testing process uses the Matlab 2018b tool. The result is a prediction with the training and testing process produces actual ouput as the target achieved.]]>
<![CDATA[Analisis Penerapan Neural Network dalam Memprediksi Produksi Bijih Nikel di Indonesia ]]>
<![CDATA[Muhammad Edya Rosadi]]>;
<![CDATA[Dian Agustini]]>;
<![CDATA[Muthia Farida]]>;
<![CDATA[Dila Dwi Anjani]]>
<![CDATA[ Brahmana : Jurnal Penerapan Kecerdasan Buatan Vol 4, No 1 (2022): Edisi Desember ]]>
Publisher : <![CDATA[LPPM STIKOM Tunas Bangsa ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.30645/brahmana.v4i1.108
<![CDATA[Nickel ore is one of the exports from the mining subsector. About 72% of the world's nickel resources are found in lateritic nickel deposits, with approximately 15.8% of these deposits located in Indonesia. Nickel is currently one of the most discussed subjects in the world. As an essential component in the creation of batteries for electric vehicles, nickel is pushing changes in energy consumption. Managing nickel ore output in Indonesia is prudent in light of the government's efforts to increase national development, investment, employment, mining downstream, and export demands. To satisfy domestic and international demand, it is essential to examine nickel ore output. Consequently, an investigation is required to forecast nickel ore production. The dataset utilized is from the Central Bureau of Statistics's Publication of Non-Oil and Gas Mining Statistics for 2017-2020. This study employs a backpropagation network with an artificial neural network. The procedure is carried out by separating training data and testing data to choose the most accurate architectural model, which is subsequently utilized as a predictive model. The architectural models to be utilized with Matlab 6.1 are 2-45-1; 2-60-1; 2-75-80-1; 2-85-1; and 2-100-1. From a series of tests, it was determined that the best architectural model was 2-45-1 with a Mean Square Error of 0.00099549, epoch 335, and an accuracy of one hundred percent. This model was then utilized to create predictions]]>
<![CDATA[Penerapan Metode PIECES Dalam Perbaikan Sistem Informasi Penjualan Berbasis Website Pada PT. Superspring Cabang Bandung ]]>
<![CDATA[Fresa Dwi Juniar Sofalina]]>
<![CDATA[ Brahmana : Jurnal Penerapan Kecerdasan Buatan Vol 4, No 1 (2022): Edisi Desember ]]>
Publisher : <![CDATA[LPPM STIKOM Tunas Bangsa ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.30645/brahmana.v4i1.133
<![CDATA[This research is motivated by the sales system that exists at PT. Superspring Bandung is still done semi-manually where it is felt that it is not effective enough to support business activities at PT. Superspring Bandung, known to PT. Superspring Bandung is a company engaged in GPS tracker marketing technology. Ideally, a technology company is able to implement an integrated information system so that business processes can run well. This research uses quantitative and qualitative methods. Where the data collected by questionnaire and interview techniques. Respondents in this study were divided into two, namely external respondents and also internal respondents where external respondents were taken as a sample of 136 respondents and internal parties as many as 3 respondents. The analysis used in this study uses the PIECES method where from the results of the initial analysis it can be seen that the current system is not effective enough according to the indicators of the PIECES method. So that from this analysis a design is made that can improve the system. The design is made using the waterfall method, where from the results of system improvements, it can be seen that updating the system can support the company's activities properly. This is known from the results of usability testing with the Likert method as an indicator for calculating the results. From the calculations above, it can be concluded that the PIECES method used as the basis of the research is considered effective enough for companies to find out the deficiencies of the existing and ongoing systems.]]>
<![CDATA[Mobile Application Sistem Monitoring Kegiatan Dan Keuangan Masjid ]]>
<![CDATA[Yanni Suherman]]>;
<![CDATA[Erien Nada Azandra]]>;
<![CDATA[Ahmad Fikri Fajri]]>;
<![CDATA[Khairil Hamdi]]>;
<![CDATA[Ade Zevimarino]]>
<![CDATA[ Brahmana : Jurnal Penerapan Kecerdasan Buatan Vol 4, No 1 (2022): Edisi Desember ]]>
Publisher : <![CDATA[LPPM STIKOM Tunas Bangsa ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.30645/brahmana.v4i1.135
<![CDATA[The information system currently used by the Mujahiddin Mosque is still manual with data recording still using books, so it is very vulnerable to data loss. Seeing from this situation, it is necessary to design an Android-based system application for the mosque. With this Android system, it facilitates and assists mosque administrators in carrying out daily activities in the mosque environment. The research method used is based on the stages in the system development method. The Android system to be built uses the JavaScript programming language. The contents of this system itself are about finances and activities at the Mujahiddin Mosque, both regarding prayer schedules, about mosque activities, mosque infaq data, and other report data so that they are more neatly organized. Thus, the existence of a financial management information system and mosque data activities at the Mujahiddin Mosque in Padang City can assist the mosque in carrying out its daily activities and activities so that it makes it easier for administrators to convey information to congregations, both in the form of activity and financial information. In addition, with this system there will be no more incidents of losing mosque data because it has been stored directly into the database system.]]>
<![CDATA[Penerapan Metode Single Moving Average Dalam Peramalan Persediaan Bahan Pangan ]]>
<![CDATA[Kukuh Rizqi Liyadi]]>;
<![CDATA[Heny Pratiwi]]>;
<![CDATA[Pitrasacha Aditya]]>;
<![CDATA[Muhammad Ibnu Sa’ad]]>
<![CDATA[ Brahmana : Jurnal Penerapan Kecerdasan Buatan Vol 4, No 1 (2022): Edisi Desember ]]>
Publisher : <![CDATA[LPPM STIKOM Tunas Bangsa ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.30645/brahmana.v4i1.136
<![CDATA[Forecasting is a technique that is quite widely used today and has been developed since the 19th century. In line with the development of increasingly sophisticated forecasting techniques accompanied by developments in the use of computers. Forecasting can predict or estimate what will happen in the future using certain techniques so that forecasting has received increasing attention in recent years. Web-based applications are one of the systems that support the development of computer use, therefore in this study, researchers develop web-based applications for forecasting using the Single Moving Average method. In this study, forecasting was carried out using the Single Moving Average method to find out how much food is needed in the following month based on actual data from the previous months. Based on forecasting which was carried out using actual data from December 2021 to June 2022, the results obtained in the following month, namely July 2022, were 2,901 kg.]]>
<![CDATA[Aplikasi Perancangan Sistem Penilaian (Ujian) Berbasis Web Di SMP Negeri 4 Kota Solok ]]>
<![CDATA[Rozi Meri]]>;
<![CDATA[Silvi Lestari]]>;
<![CDATA[Aidil Putra]]>
<![CDATA[ Brahmana : Jurnal Penerapan Kecerdasan Buatan Vol 4, No 1 (2022): Edisi Desember ]]>
Publisher : <![CDATA[LPPM STIKOM Tunas Bangsa ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.30645/brahmana.v4i1.137
<![CDATA[Learning is a very important element in education in Indonesia. In learning there are various kinds of strategies and methods that can be used in accordance with existing conditions. Implementation of learning strategies which include teaching, discussion, reading, assignments, presentations and evaluations. This includes online assessments with the Learning Management System (LMS). With the assessment process it is easier and more efficient. With this online/digital system (LMS), the assessment process can be carried out quickly and accurately. After students carry out the assessment, grades can immediately appear in the system. This final project aims to assist teachers and students in administering exams, especially at SMPN 4 Kota Solok. In the implementation of the exam still uses conventional methods that require quite a lot of time and effort. The role of information and communication technology that is growing rapidly, one of which is through the internet network, can be used to overcome this problem. The study conducted in this research created a website for online exam implementation. This can make it easier for teachers to upload questions that have been made for online exam implementation. The advantage of the system contained in this web is that students can immediately find out the results of the exam after ending it.]]>
<![CDATA[Analisis Pengaruh Komposisi Data Training dan Testing Terhadap Akurasi Algoritma Resilient Backpropagation (RProp) ]]>
<![CDATA[Harly Okprana]]>;
<![CDATA[Riki Winanjaya]]>
<![CDATA[ Brahmana : Jurnal Penerapan Kecerdasan Buatan Vol 4, No 1 (2022): Edisi Desember ]]>
Publisher : <![CDATA[LPPM STIKOM Tunas Bangsa ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.30645/brahmana.v4i1.138
<![CDATA[Prediction classification accuracy is a measure of success and satisfaction in predicting past data to produce accurate predictions, knowing how precise a classification pattern predicts class data from future data. In practice, artificial neural networks test the accuracy of a classification pattern using data testing, while to find the pattern itself, use training data. Errors in determining the composition of the presentation of training and testing data can affect the accuracy value obtained, therefore the distribution of the presentation of the amount of data from a dataset is one of the determining factors for the amount of accuracy. This study uses a dataset of Michigan Computer English Course students in 2018-2019 using the Resilient Backpropagation (RProp) method. The data processed was 100 student data for 2018-2019. By dividing the composition of 25% training data with 75% data testing with an accuracy value of 99.25% while dividing 50% training data with 50% data testing with an accuracy value of 100% as well as dividing 75% training data with 25% data testing with a value 100% accuracy.]]>
<![CDATA[Perbandingan Algoritma Adaline Berdasarkan Pola Input Data Dan Aktivasi Output Untuk Prediksi Data ]]>
<![CDATA[Donni Nasution]]>;
<![CDATA[Darmeli Nasution]]>;
<![CDATA[S Solikhun]]>
<![CDATA[ Brahmana : Jurnal Penerapan Kecerdasan Buatan Vol 4, No 1 (2022): Edisi Desember ]]>
Publisher : <![CDATA[LPPM STIKOM Tunas Bangsa ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.30645/brahmana.v4i1.139
<![CDATA[Adaline is a single-layer supervised learning algorithm where the input layer is directly related to the output layer. Adaline learning uses the delta rule, which adjusts the weights to reduce the difference between network inputs to the desired output and output units. The main problem of this study is to find an alternative to the Adaline algorithm for predicting stroke with seven symptom attributes. This study seeks the best Adaline algorithm performance by comparing four forms of input and output activation patterns. The test results show the results of the same accuracy that is equal to 100%; the same epoch, namely one epoch, and the average weight change is different. The Adaline algorithm can predict stroke well with 100% accuracy.]]>
<![CDATA[Mikrotik Optimization and Pc Cloning on Computer Networks with Using UBIQU Satellite Internet Vsat Broadband ]]>
<![CDATA[S Sumarlin]]>;
<![CDATA[Muhammad Zarlis]]>;
<![CDATA[S Suherman]]>;
<![CDATA[Syahril Efendi]]>
<![CDATA[ Brahmana : Jurnal Penerapan Kecerdasan Buatan Vol 4, No 1A (2022): Edisi Desember (Spesial Issue) ]]>
Publisher : <![CDATA[LPPM STIKOM Tunas Bangsa ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.30645/brahmana.v4i1A.140
<![CDATA[The internet is currently growing rapidly. The need and rapid development in the use of the Internet network requires a balance in the provision of Internet facilities. Internet standard services are the continuity of the connectivity of the Internet. Connections from the Internet are required to always be awake under any conditions, but not always connectivity will run smoothly, many obstacles or disturbances are encountered so that the connection does not run smoothly. One effective method that can be done is to apply the PCQ method in dividing the amount of internet bandwidth by paying attention to the distribution pattern of bandwidth speed based on the number of active users at a certain time. The subjects taken in this study were users who were active during the active period of students and other users in the Hotspot network at the Indonesian Institute of Technology and Business. The method used is a case study method based on the level of activity of the hotspot network on campus. Comparison between the amount of bandwidth for the number of users between the hours of high activity in the morning until the afternoon, and the hours of low activity in the afternoon until the evening. The PCQ method will divide the bandwidth based on the number of active users on the hotspot with the same amount. The design is continued by applying the PCQ method using Miktrotik Routerboard, testing is carried out by technical testing by testing the amount of bandwidth with division based on the number of active users and the same results are obtained.]]>
<![CDATA[Algoritma Backpropagation Metode Powell-Beale Untuk Prediksi Penyakit Stroke ]]>
<![CDATA[Verdi Yasin]]>;
<![CDATA[Ifan Junaedi]]>
<![CDATA[ Brahmana : Jurnal Penerapan Kecerdasan Buatan Vol 4, No 1A (2022): Edisi Desember (Spesial Issue) ]]>
Publisher : <![CDATA[LPPM STIKOM Tunas Bangsa ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.30645/brahmana.v4i1A.156
<![CDATA[Stroke is a dangerous disease, a stroke is caused by a blockage of blood to the brain or insufficient blood supply to the brain. Stroke can cause brain damage, long term disability and death. The purpose of this study focuses on stroke prediction using the conjugate gradient Powell-Beale backpropagation algorithm. This stroke prediction data is taken from kaggle which consists of 5110 records. The attributes used to predict stroke consist of 7 attributes, namely age, hypertension, heart disease, marital status, average blood sugar, BMI, and smoking status. The results of this study are the prediction of stroke with MSE training and testing = 0. With a 7-2-1 architecture.]]>
