Rabit : Jurnal Teknologi dan Sistem Informasi Univrab

MONITORING ADMINSTRASI JARINGAN DENGAN MIKROTIK DAN TELEGRAM BOT PADA INTERNET SERVICE PROVIDER

Taufik Rahman, Iqbaluz Zain Nibras, Sumarna Sumarna — Mon, 08 Jul 2024 19:29:46 +0700

Manajemen jaringan menggunakan perangkat MikroTik di Lajur Net masih memiliki kekurangan. Walaupun akses internet tersedia sepanjang waktu, keterbatasan jumlah karyawan membuat pengelola sulit mengakses komputer untuk mengelola MikroTik. Administrator berharap bisa mengatur jaringan MikroTik secara remote melalui perangkat akhir. Saat ini, akses MikroTik secara daring memerlukan biaya operasional yang tinggi untuk menyewa IP publik dan layanan penyediaan aplikasi manajemen berbasis web. Penelitian ini bertujuan untuk memanfaatkan Telegram Bot dalam pengelolaan dan pemantauan jaringan yang dapat diakses dari jarak jauh melalui internet. Penggunaan Telegram Bot untuk pengelolaan jaringan di Lajur Net dapat diimplementasikan dengan sukses. Penerapan Telegram Bot mencakup pengelolaan interface, hotspot, ICMP, dan pemantauan sumber daya perangkat RouterBoard MikroTik di jaringan Lajur Net. Telegram Bot dapat diakses dari jarak jauh melalui jaringan internet dan berjalan dengan baik pada jaringan Lajur Net. Penerapan Telegram Bot meliputi administrasi interface, hotspot, ping, serta monitoring sumber daya perangkat RouterBoard MikroTik. Fitur tambahan berupa firewall memungkinkan administrator jaringan untuk menambah dan menghapus filter rules dalam sistem keamanan jaringan Lajur Net. Selain itu, Telegram Bot memanfaatkan fitur tombol pada pesan Telegram untuk mempermudah penggunaan tanpa harus mengetikkan semua perintah. Dengan demikian, Telegram Bot menawarkan solusi yang efisien dan efektif dalam manajemen jaringan MikroTik di Lajur Net, mengurangi kebutuhan akan IP publik dan aplikasi manajemen berbasis web yang mahal. Hal ini membuktikan bahwa penggunaan teknologi chatbot dapat memberikan manfaat signifikan dalam bidang administrasi jaringan, khususnya bagi penyedia layanan internet dengan sumber daya terbatas. Melalui penelitian ini, diharapkan implementasi Telegram Bot dapat dijadikan acuan untuk pengembangan manajemen jaringan yang lebih inovatif dan efisien di masa mendatang.

IMPLEMENTASI CONVOLUTIONAL NEURAL NETWORK DENGAN PRE-TRAINED MODEL MOBILENETV2 UNTUK DETEKSI KOLESTEROL

Indah Sri Lestari, Jumadi Jumadi, Nur Lukman — Mon, 08 Jul 2024 19:30:16 +0700

One of the main risks for heart disease and stroke is cholesterol. Cholesterol is a type of fat produced primarily by the liver and absorbed in small amounts from food. The ideal cholesterol level in the human body should be less than 200 mg/dl. One way to check cholesterol levels is through a blood sugar test that requires the patient to fast for 10 to 12 hours. Given the dangers posed by high cholesterol levels, there is a need for an early, practical screening method to detect high cholesterol levels in the human body. Iridology is an analysis of the iris of the eye to detect health conditions and show the relationship between iris patterns and cholesterol levels. The iris has its own uniqueness because it can record the condition of all organs, body structures, and psychological states. Therefore, iridology can be an alternative medical analysis. This study proposes the use of a convolutional neural network algorithm using a pre-trained MobileNetV2 model. The iris image dataset used consists of 200 images classified into two classes: normal eye images and cholesterol eye images. The results of the study show that the proposed model can achieve an accuracy of 95%. These results indicate that this model has great potential as a practical and cost-effective tool for detecting cholesterol. Further research is needed with larger datasets to improve accuracy and validity.

PEMANFAATAN STFT DAN CNN DALAM PENGOLAHAN DATA SUARA UNTUK MENGKLASIFIKASIKAN SUARA BATUK

Indri Nurfiani, Jumadi Jumadi, Muhammad Deden Firdaus — Mon, 08 Jul 2024 19:38:34 +0700

This research aims to develop an automatic cough sound evaluation system to improve the accuracy of respiratory disease diagnosis. In this study, the Short-Time Fourier Transform (STFT) and Convolutional Neural Network (CNN) methods were used to classify cough sounds into dry and wet coughs. The Naïve Bayes model was then used to identify respiratory diseases based on the cough classification results. Testing was conducted using the available cough sound dataset, resulting in a cough classification accuracy of 82% and a respiratory disease identification accuracy using Naïve Bayes of 71.43%. The evaluation results indicate that the developed system can accurately classify cough types and identify diseases. This system has the potential to enhance the prevention and management of respiratory diseases in resource-limited areas and can be a significant tool in medical practice for faster and more accurate diagnoses. Furthermore, this research opens opportunities for further development in disease detection and diagnosis technology through sound analysis, providing wide-ranging benefits for society and the healthcare sector.

ANALISA DAN DESAIN KONTEN INSTAGRAM YANG MENARIK DAN BERDAMPAK POSITIF PADA INTERAKSI PENGGUNA

Sri Mulyani, Purwatiningtyas Purwatiningtyas, Yunus Anis — Mon, 08 Jul 2024 20:50:24 +0700

As one of the most popular social media platforms today, Instagram has more than one billion monthly active users. This is a great opportunity for users to interact with each other, share content, and build rapport. However, to achieve this, the right strategy is needed to create interesting content according to user interests. In line with the aim of this research, to obtain interesting content that has a positive impact on user interaction, researchers use the Design Thinking method which involves users and stakeholders in Instagram accounts. This research was conducted by including three content designs that provide the best insight from more than sixty Instagram content designs posted on accounts in the period 5 June to 6 September 2023 on the Graphic Multimedia Engineering Technology (TRMG) study account @trmg_unisbank which was used as an object. From the analysis and design of Instagram content, it managed to reach 20,275 followers with 80 accounts and 20,195 non-follower accounts. This means that attractive content design can have a positive impact on user engagement. This content is built by exploring the target audience, understanding user interests, using images and design, including interesting captions and relevant hashtags, and conveying useful information (education).

SISTEM PAKAR DIAGNOSA POST TRAUMATIC STRESS DISORDER DENGAN METODE FORWARD CHAINING

Mochamad Rizky Zaldi, Seng Hansun — Mon, 08 Jul 2024 20:58:52 +0700

Post traumatic stress disorder (PTSD) is a disorder experienced by someone due to surprising, scary, or dangerous events they had. Unfortunately, there is minimal knowledge about this disorder among Indonesian people so that many of them do not know what symptoms they have experienced and the effects from the disorder. The aim of this study is to build an expert system for early diagnose of this PTSD. Forward chaining method is utilized within the expert system to collect the symptoms experienced by users, processing them based on knowledge rules until a decision or conclusion achieved. The expert system was developed following the Expert System Development Life Cycle (ESDLC) and the built system usability level was measured by using the USE questionnaire. The system was then tested together with an expert that showed the agreement level of the system with the expert. From the Usability level measurement, the system got a score of 86.67% which show the system is useful for early diagnose of PTSD. Hopefully, this expert system could become a solution in giving ease of accessibility for people to get an early diagnose result of PTSD.

ANALISIS FAKTOR YANG MEMPENGARUHI PEMILIHAN GUBERNUR DAERAH KHUSUS JAKARTA MENGGUNAKAN ALGORITMA NAIVE BAYES DAN REGRESI LOGISTIK

Yoga Nanda Khoiril Umat, Dhiaz Rusyda Nafsyi, Diana Kusumaningsih, Lukman Hakim — Mon, 08 Jul 2024 21:10:05 +0700

The election of the Governor and Deputy Governor of the Special Region of Jakarta (DKJ) in 2024 involves the community in determining leaders for the 2024-2029 period. This research analyzes the factors influencing voter decisions using the Naive Bayes algorithm and Logistic Regression. Survey data was collected via a Google Form questionnaire from Jakarta residents aged 17-71. The analysis process involves several stages: problem identification, literature study, data collection, preprocessing, and dividing the data into training and test data. The Naive Bayes algorithm is used to predict classification parameters based on a candidate's education, popularity, and track record, while Logistic Regression predicts factors that influence voter decisions. Naive Bayes shows high accuracy with advantages in speed and processing large data, while Logistic Regression shows strength in binary and multinomial classification analysis. The research results show that the track record factor significantly influences voter decisions. Naive Bayes prediction accuracy reached 85.00% and Logistic Regression 80.00%. The analysis results also reveal that the candidate's popularity and education factors rationally influence voter decisions, although not as strong as the track record. In addition, using these two algorithms provides a comprehensive understanding of voter behavior in Jakarta. Based on these results, governor and deputy governor candidates should also focus on improving their track record and popularity to increase their chances of being elected.

PENERAPAN JARINGAN SARAF TIRUAN DENGAN ALGORITMA PERCEPTRON PADA DETEKSI PENYAKIT BAWANG MERAH

Mon, 08 Jul 2024 21:15:54 +0700

Penelitian ini difokuskan pada kemajuan sistem deteksi penyakit bawang merah (Allium ascalonicum) menggunakan Jaringan Saraf Tiruan (JST) dengan Algoritma Perceptron. Bawang merupakan komoditas hortikultura penting dengan kepentingan ekonomi yang substansif di Indonesia. Namun, produksi bawang merah sering terhambat oleh berbagai penyakit yang dapat menyebabkan kerugian signifikan. Identifikasi tepat waktu dan pengelolaan yang efektif terhadap penyakit tersebut sangat penting untuk mengurangi dampak buruknya. Sistem pakar yang memanfaatkan kecerdasan buatan, khususnya JST, telah terbukti efektif dalam deteksi penyakit tanaman. JST meniru cara kerja otak manusia dalam mengenali pola dan membuat keputusan berdasarkan data yang diperoleh. Penelitian ini menggunakan data gejala penyakit bawang merah yang terdiri dari 13 gejala dan 7 jenis penyakit. Setiap gejala dan penyakit dihubungkan dalam basis pengetahuan yang kemudian dianalisis menggunakan Algoritma Perceptron. Perceptron adalah algoritma pembelajaran terawasi yang digunakan untuk klasifikasi dan pengenalan pola. Proses pelatihan melibatkan inisialisasi bobot dan bias, perhitungan respon output, dan penyesuaian bobot jika terjadi kesalahan. Hasil akhir dari sistem ini adalah identifikasi jenis penyakit yang menyerang bawang merah berdasarkan gejala yang diamati. Implementasi sistem ini mencakup tampilan home, tampilan diagnosa, dan tampilan hasil diagnosa. Sistem ini memungkinkan pengguna untuk melakukan diagnosa penyakit bawang merah dengan menjawab pertanyaan terkait gejala yang diamati. Hasil diagnosa akan menampilkan jenis penyakit dan solusi yang disarankan. Penelitian ini menunjukkan bahwa Algoritma Perceptron dapat diterapkan secara efektif dalam sistem deteksi penyakit bawang merah, membantu petani dalam mengidentifikasi dan mengelola penyakit dengan lebih cepat dan akurat.

PREDIKSI JUMLAH KEDATANGAN WISATAWAN MANCANEGARA DI INDONESIA BERDASARKAN PINTU MASUK KEDATANGAN UDARA

Arya Prayuda, Irfan Pratama — Mon, 08 Jul 2024 21:22:50 +0700

Indonesia has diversity and natural wealth that attracts tourism in Indonesia. Tourism is one of the industries that provides the highest foreign exchange for the country because it has a positive impact. However, the existence of COVID-19 has resulted in a decrease in the number of visits due to restrictions on foreign tourists. From January to November 2021, there was a drastic decrease of 61.82% in the number of foreign tourist visits compared to the same period in 2020. In addition to COVID-19, as well as support in building facilities that support the increase in the number of foreign tourists. From these conditions, predictions are needed that are used as a basis for planning and helping decision making. The purpose of this study is to develop a more accurate prediction model in similar studies using the same data in predicting foreign tourist arrivals in Indonesia through air entrances using the XGBoost, Random Forest, and Catboost methods by focusing on the accuracy evaluation results metrics RMSE, MAE, and MAPE and making predictions for the next 12 months. The dataset used is taken from the Central Statistics Agency (BPS), namely data on foreign tourist arrivals based on the arrival entrance in the period January 2017 to November 2021. The data used are time series and non-stationary. From the research results, it can be seen based on the accuracy evaluation results that the XGBoost model of this study gets better accuracy evaluation results than the other two models by getting the results of the RMSE accuracy evaluation value of 671935.2, MAE 648139.1, and MAPE 20985.35. The XGBoost model is better with a smaller accuracy error value than the Random Forest model, Catboost, and similar research using the ARIMA method with an RMSE value of 779670.7, MAE 749030.4, and MAPE 23196.45.

PENANGANAN MISSING VALUES DAN PREDIKSI DATA TIMBUNAN SAMPAH BERBASIS MACHINE LEARNING

Anisa Widianti, Irfan Pratama — Mon, 08 Jul 2024 21:30:35 +0700

The issue of increasing waste due to the growing population and human activities presents a serious challenge in waste management in Central Java. One of the main obstacles in waste prediction research is the prevalence of missing data, which can reduce the accuracy of predictive models. This study employs three methods to handle missing values: Mean Imputation, Interpolation, and KNN Imputer. Once the missing values are filled using these methods, the next step is to calculate the prediction values. The study utilizes three predictive models: Random Forest, Gradient Boosting, and KNN. The results indicate that with Mean Imputation, the Random Forest model shows the best performance with an RMSE of 0.349. When using Interpolation for missing values, the Gradient Boosting model becomes the best choice with an RMSE of 0.543. Meanwhile, with KNN Imputer, the Gradient Boosting model again performs the best with an RMSE of 0.188. Based on this research, the most effective approach is using KNN Imputer for handling missing values in conjunction with the Gradient Boosting model. This combination provides the lowest RMSE for similar datasets.

PENERAPAN METODE ROTOSCOPING DAN FRAME ELIMINATION PADA ANIMASI 2D TARI ZAPIN

Mon, 08 Jul 2024 21:42:38 +0700

The growth of the animation industry in Indonesia is predicted to continue to grow. This can be seen from the development of animation in Indonesia, which is dominated by 2-dimensional and 3-dimensional animation in recent years, as well as being marked by the large number of animators from Indonesia who are successful on the international stage. Generally, animation is applied to character movements, whether simple movements such as walking, to complex movements such as fighting. In its application, various methods can be used as long as they can create animated movements that match the original movements. Generally this is done successfully because in complex movements, the speed of the movement seems to obscure the animation details of the movement. However, this does not apply to one type of movement that is very rarely used in animation, this movement is dancing. The reason why dance movements are rarely animated is because dance is a series of movements that have low speed, so that the details in each change in movement in one dance can produce hundreds or even thousands of frames when converted into animation. Visualizing dance art with animated visuals is not an easy thing, because there are many frames that have to be created so that the dance looks natural. Moreover, using animation techniques such as frame by frame and cutout, of course, actually increases the processing time and quantity of images produced. Actually, apart from frame by frame and cutout, there are still other animation techniques that can be applied to overcome the problem of many frames in dance movements. One solution is to visualize dance in animated form. Namely the rotoscoping technique which can bridge dance with animation. This rotoscoping imitates every change in movement in dance, and converts it into a frame. However, if you only use this technique, the resulting image will still be too large, so one more technique needs to be added, namely frame elimination. For this reason, the author wants to apply the rotoscoping technique in making Zapin dance animations. To find out whether dance art can be visualized well using rotoscoping and Frame Elimination techniques.

PENERAPAN TEKNIK PENETRATION TESTING TERHADAP CROSS SITE SCRIPTING (XSS) DALAM PENGEMBANGAN WEBSITE

Ahmad Alfian Chandra, Ahmad Turmudi Zy, Agung Nugroho — Tue, 09 Jul 2024 06:08:38 +0700

The increasing use of websites in various aspects of daily life has led to an urgent need to ensure the security of the information presented. One of the significant threats in website security is Cross-Site Scripting (XSS), where an attacker inserts malicious code into a web page to be executed by the user. This research aims to apply penetration testing techniques as a method to detect and resolve XSS vulnerabilities in website development. The research was conducted through three stages: installation of software to support penetration testing, execution of penetration testing using OWASP ZAP to identify vulnerabilities, and evaluation and implementation of solutions to address the vulnerabilities found. The results show that the implementation of the htmlspecialchars function in PHP is effective in preventing the execution of malicious scripts, thereby reducing the risk of XSS attacks. In addition, penetration testing techniques proved to be an effective method in identifying and mitigating security risks in web applications. Thus, this research emphasizes the importance of thorough security testing and implementation of appropriate preventive measures to maintain the integrity and user trust of web applications.

PERANCANGAN SISTEM INFORMASI PENGGAJIAN KARYAWAN BERBASIS WEBSITE DI PERUSAHAAN X MENGGUNAKAN METODE PROTOTYPE

Deri Ramadiyan Hidayat, Dani Hamdani — Tue, 09 Jul 2024 06:18:01 +0700

Company X is a growing company along with increasing customer demand and employee needs, which means a computerized system is needed to reduce errors in salary calculations. Currently the company is still less effective in calculating employee salaries because it still calculates, pays salaries and makes reports manually with methods in data collection, namely using interviews with company owners, conducting field observations and also questionnaires used for assessment of system design, the design method uses prototypes which have stages, namely listening to customers or collecting needs, Design and Test in assessment The trial uses a System Usability Scale (SUS) assessment of the respondent's assessment data, then calculates using the formula from the System Usability Scale, the smallest value is 78 and the largest value is 90 from the depalan respondent by answering 10 questions which has an average score of 86 from the qualifications in the table above "Very Feasible" the user interface must be designed so that it is easy to use and easy to understand so that payroll data input errors can be avoided. The use of autofill features and real-time data validation can help reduce input errors. And to create an effective user interface, it is necessary to understand user needs and expectations well. This can be achieved by involving users during the design and testing process, such as surveys, interviews, and prototype trials.

OPTIMALISASI PERANCANGAN APLIKASI MANAJEMEN DATA KOLOMAN, DI DESA PULAU MANDANGIN SAMPANG – MADURA BERBASIS WEBSITE

Tue, 09 Jul 2024 06:32:13 +0700

The rapid development of information technology has had a significant influence on various aspects of life, including data management in traditional activities such as koloman in Pulau Mandangin Village, Sampang, Madura. The manual data management that residents still apply creates difficulties in the process of searching and managing koloman data. This causes the time required to enter and search for information to be longer, and increases the risk of inaccuracies. Additionally, physical storage of information can also pose security and support issues, as information can be lost or damaged. As a result, data management becomes inefficient and people have difficulty finding the information they need quickly and accurately. To overcome these problems, this research proposes an innovative and sustainable solution for designing a web-based koloman data management application. This application is designed to facilitate centralized data processing, automate data input, and provide real-time access to information. Thus, it is hoped that this application can increase the accuracy, speed and transparency of koloman data management, as well as provide significant benefits for the community in managing koloman data more effectively and efficiently. The application of web-based technology also opens up opportunities to increase the capacity of local communities in utilizing information technology.

PENGGUNAAN ALGORITMA GEOSHASHING DALAM PENCARIAN LETAK INFORMASI GEOGRAFIS PENGGANTI KOOORDINAT LATITUDE DAN LONGITUDE

Fajar Firdaus, Asep Id Hadiana, Edvin Ramadhan — Tue, 09 Jul 2024 06:47:37 +0700

Dalam era digital saat pencarian data geografis menjadi semakin penting dalam aplikasi lokasi, termasuk pemetaan, navigasi, dan analisis lingkungan tempat dalam suatu lokasi. algoritma geohashing menawarkan pendekatan yang efisien untuk pengkodekan dan menyimpan data geografis dalam format string,String ini menggunakan alfanumerik yang dihasilkan oleh geohashing dari angka dan huruf dalam pencarian titik lokasi dengan koordinat, hal ini juga dapat menawarkan lapisan akses dari data lokasi yang sensitif. Penelitian ini bertujuan untuk mengeksplorasi penerapan algoritma geohashing dalam manajemen data geografis. algoritma geohashing dalam pencarian letak informasi geografis sebagai alternatif pengganti koordinat latitude dan longitude. Secara khusus, penelitian ini menganalisis efisiensi geohashing dalam kompresi data, akurasi lokalisasi penggunaan geohashing memungkinkan pengelompokan data geografis yang lebih baik, dan mendukung pencarian data spasial seperti pencarian letak lokasi dan pencarian titik koordinat dalam wilayah tertentu. algoritma geohashing diimplementasikan dalam data geografis dan diuji untuk berbagai skenario pencarian. Dari hasil penelitian menunjukkan bahwa geohashing dapat meningkatkan kecepatan pencarian informasi geografis secara signifikan algoritma geohashing membuktikan tindakan sebagai alat yang efektif dalam manajemen dan pencarian data geografis,pada pencarian data koordinat lokasi ini juga dapat memberikan alternatif untuk mendapatkan latitude dan longitude dari suatu lokasi. Hasil penelitian menunjukkan bahwa geohashing dapat meningkatkan kecepatan pencarian informasi geografis secara signifikan, terutama dalam menjangkau suatu lokasi dengan level grid kotak data yang kecil dan besar sesuai karakter string yang dimasukan. Penggunaan geohashing tidak hanya meningkatkan efisiensi dan kecepatan pencarian tetapi juga memberikan manfaat tambahan dalam hal pengelolaan dan perlindungan data geografis.algoritma geohashing adalah metode yang digunakan untuk mengkodekan koordinat geografis yang menghasilkan lokasi acak.Pengujian pada sistem ini dilakukan dengan cara melakukan akses geografis kepada API aplikasi lokasi yang sudah dibuat sesuai permintaan yang diminta untuk mendapatkan informasi data geografis dan titik lokasi.

ANALISIS DAN DESAIN SISTEM PENJUALAN KONSINYASI MENGGUNAKAN PENDEKATAN BERORIENTASI OBJEK LIGHTWEIGHT RATIONAL UNIFIED PROCESS

Mohammad Fajar, Richla Rahmatika — Wed, 10 Jul 2024 09:53:31 +0700

The aim of this study is to apply Lightweight Rational Unified Process (RUP) approach in modeling consignment sales information systems. Data collection is performed through literature studies, interviews and evaluation of modeling results systems. The application of the Lightweight RUP method in this research is carried out by following each stage of the approach, starting from requirements stage, analysis stage, then design stage, and the implementation stage where the connection between one stage and another can be maintained through the artifacts produced. In the requirements stage, usecase is made based on the requirements table which contains system features based on literature studies related to sales information systems and through interviews of the potential users. Furthermore, the usecase is the basis for making a number of diagrams at the analysis stage, namely activity diagrams and sequence diagrams. At the design stage, platform selection needs to be done first, so that the class diagram that will be designed to adjust to the platform chosen. In addition, at the final stage of the design, the author adds the evaluation process to the completeness of the class diagram before translated them into program code. Evaluation at the final stage of the design shows that the created class diagram has fulfilled all aspects of completeness so that it is possible to be mapped to the final target platform selected, namely the Java programming language. While the evaluation of generated program code on the Java platform shows that the cashier will able to use the desktop application to make sales transactions. As for the Android platform-based application can be used by the consignor to see the information on the status of the sold items (cakes).

PENERAPAN HOUGH TRANSFORM, CONNECTED COMPONENT LABELING DAN TEMPLATE MATCHING UNTUK PENGENALAN KARAKTER PLAT KENDARAAN

Fredicia Fredicia, Gardyan Santoso — Wed, 10 Jul 2024 11:51:33 +0700

Vehicle license plates are the identities of motor vehicles that are often used for record-keeping when using paid parking lots. Recording vehicle license plates requires the application of information technology so that parking service administration is more efficient and free from recording errors. Information technology offers the application of OCR (Optical Character Recognition) methods to help the process of recording characters found on vehicle license plates. OCR has many techniques that can help the character recognition process, starting from the initial image capture process to the recognition stage. The initial image stage after being captured is to perform the initial image processing process by converting the image, determining the vehicle license plate area in the vehicle image, performing character labeling and segmentation on the plate image, and performing character recognition. The approach to character recognition on vehicle license plates that will be used is Hough Transform for determining the vehicle license plate area, character labeling and segmentation with Connected Component Labeling, and Template Matching as a character recognition method. The vehicle license plate character recognition model was tested with a scenario of 25 images captured directly from the parking lot. The test results produced Hough Transform and Connected Component Labeling could determine the vehicle license plate area and perform character labeling and segmentation on the vehicle license plate. Whereas in the template matching stage, the character recognition accuracy on 25 vehicle license plates was 94%. Other findings in this study are that the lighting conditions of the image environment, the number of images that become the database, and the condition of the paint of motor vehicle license plates can improve accuracy.

PENERAPAN FUZZY ASSOCIATIVE MEMORY UNTUK MENGUKUR TINGKAT KEPUASAN PELANGGAN TERHADAP PELAYANAN PADA TOKO KOPIMA

Nur Wakhidah, Ira Nurul Febriani — Wed, 10 Jul 2024 17:02:31 +0700

One of the advantages of information technology in the field of business is that it can help in measuring the level of customer satisfaction. Because measuring customer satisfaction is an important element in providing better, more efficient and more effective services. KOPIMA stores, which is a student cooperative, is experiencing unstable sales problems, this is the result of customer dissatisfaction. There is a system used to measure the level of customer satisfaction with services at KOPIMA stores to find out what are the expectations of customer complaints and how to evaluate service at KOPIMA stores. To measure the level of customer satisfaction with services at KOPIMA stores, the Fuzzy Associative Memory (FAM) method is used. FAM is a flexible decision-making method, which is a fuzzy system that maps fuzzy sets to other fuzzy sets. The output obtained is a web-based system, with the results of the FAM method that can be implemented to calculate customer satisfaction into the rating category, namely 8% satisfied and 92% very satisfied in the November 2022 assessment and December 2022 assessment, namely 4% satisfied and 96% very satisfied . So it can be concluded, customer satisfaction has increased with the assessment criteria being very satisfied with the service at the KOPIMA store.

KLASIFIKASI HASIL MRI TUMOR OTAK DENGAN EKTRAKSI FITUR GRAY LEVEL CO-OCCURANCE MATRIX (GLCM)

Fari Katul Fikriah, Amelia Devi Putri Ariyanto, Arif Fitra Setyawan — Thu, 11 Jul 2024 08:53:24 +0700

Bagian penting dari tubuh adalah otak yang mana menjadi sumber dari semua alat tubuh yang terletak dalam rongga tengkorak, tumor otak merupakan salah satu penyakit yang dapat menyerangnya. Pendeteksian tumor otak adalah salah satu aspek yang dinilai penting dalam diagnosa medis. Pada penelitian ini memiliki tujuan melakukan implementasi ekstraksi fitur GLCM (Gray Level Co-occurence Matrix) pada citra MRI tumor otak serta mencari performa algoritma yang paling baik dari deteksi tumor otak menggunakan citra MRI ini. Data yang dipakai pada penelitian ini merupakan data public yang berasal dari kaggle.com. Proses ekstraksi fitur pada citra digunakan pada penelitian ini GLCM yang mana memiliki fungsi menghitung frekuensi dari nilai intensitas piksel yang berjarak antar citra dengan menggunakan parameter 0^o, 45^o, 90^o, 135^o. Tahap selanjutnya pada penelitian ini adalah dengan melakukan langkah preprocessing dengan selanjutnya mencari nilai klasifikasi dari hasil MRI menggunakan algoritma Naïve Bayes, C4.5 dan Neural Network. Hasil yang didapatkan memperlihatkan bahwa Naïve Bayes memiliki performa algoritma paling baik dibandingkan C4.5 dan Neural Network yaitu dengan akurasi algoritma Naïve Bayes sebesar 96.8%, sedangkan untuk algoritma C4.5 sebesar 41.5% dan Neural Network sebesar 38.25%. selain hal tersebut pada penelitian ini membuktikan bahwa dengan ekstraksi fitur GLCM terbukti efektif dalam menangkap informasi tekstur dari citra MRI yang sangat penting pada klasifikasi tumor otak.

SISTEM PENDUKUNG KEPUTUSAN PROMOSI KENAIKAN JABATAN MENGGUNAKAN METODE ANALYTICAL HIERARCY PROCESS

Fri, 12 Jul 2024 15:12:03 +0700

Promosi kenaikan jabatan bagi perusahaan merupakan rutinitas dan sebagai salah satu bentuk penghargaan kemampuan karyawan selama bekerja. Selama ini promosi jabatan di PT. XYZ masih berdasarkan ikatan keluarga, tingkat pendidikan dan kedekatan personal karyawan dengan pimpinan. Sering sekali terjadi permasalahan dalam perekrutan promosi jabatan yang berakibat terjadinya kecemburuan diantara karyawan. Masih adanya sikap like dislike yang memungkinkan karyawan yang memiliki kompetensi merasa tidak dihargai. Belum adanya sistem yang mengakomodir keputusan promosi jabatan sehingga penilaian promosi jabatan belum sesuai dengan kompetensi yang dimiliki oleh karyawan termasuk tanggung jawab, disiplin, kerjasama dan perilaku/atitude karyawan. Metode yang digunakan Analytical Hierarcy Proses (AHP) dimana keputusan diambil berdasarkan kriteria dan alternatif yang telah ditentukan. Dilakukan proses Focus Group Discussion (FGD) dalam proses penilaian karyawan yang dipromosikan jabatannya. Sistem ini dibuat untuk mengatasi permasalahan yang ada dan membantu pimpinan dalam menentukan keputusan promosi jabatan agar sesuai dengan kebutuhan perusahaan berdasarkan tanggung jawab, disiplin, kerjasama dan perilaku atau attitude karyawan.

RANCANG BANGUN ALAT UKUR KADAR GULA DALAM DARAH NON INVASIVE MENGGUNAKAN ESP32 DAN BLYNK

Vadel Amanika, Sunanto S.Kom,M.Kom — Fri, 12 Jul 2024 15:30:33 +0700

The compatibility of ESP32 microcontrollers in Internet of Things (IoT) implementations covers almost all fields. Has the potential to monitor blood sugar levels non-invasively with the presence of Fotodioda Sensors and Infared LED lights and fingers as measurement objects without the need to injure patients or invasively in the aim to improve public health services, especially in an effort to prevent diabetes mellitus. ESP32 microcontroller also has compatibility in the use of Blynk which can be used to be a medium to monitor blood sugar levels and can be displayed on a computer monitor or smartphone by opening Blynk private, also with the utilization of Light intensity using fotodioda sensors and LED lights will be designed Non-Invasive Blood Sugar Measuring Instruments Using Prototyping Methods by utilizing ESP32 and Blynk and conducted experiments to see performance and comparison of accuracy with Glucometers in general and utilization in applications in the Health sector. In the prototype experiment of the Blood Glucose Measurement Device, the light absorption calculation produced by the combination of photodiode sensors and LED lamps can be utilized for measuring blood glucose levels without the need for direct blood sampling from the user. The implementation is straightforward and includes a monitoring system using Blynk, facilitating the process of blood glucose monitoring via both smartphones and users' computers. The overall accuracy result of the non-invasive blood glucose measurement device prototype reached 95% based on multiple conducted experiments.

EVALUASI APLIKASI MOBILE DESAIN MENGGUNAKAN METODE NIELSEN’S ATTRIBUTES OF USABILITY

Khadijah Hulliya, Aliza Fatha Amanda — Mon, 15 Jul 2024 14:01:03 +0700

To achieve good software standards, software quality testing is required, including usability. One of the factors that determine software quality is ease of use. In this study, the authors conducted research on the Canva application. This application has been downloaded 100 million times by users on Google Playstore. In this study, the authors used the usability testing method. The data used in this study were obtained through questionnaires and literature studies. The purpose of this research is to find out the results of the usability evaluation of the Canva application. Data was obtained through literature research and Nielsen's questionnaire on usability features. The use of Nielsen's Usability Attributes questionnaire method, which has five indicators: Learnability (easy to learn), Efficiency (efficient), Memorability (easy to remember), Errors (failure), and Satisfaction (satisfaction), the purpose of this study is to determine the results of usability evaluation on the Canva application. Testing with the NAU method gives an average result of 4.02 (Agree) from a rating scale of 5 (Strongly Agree). That is, users agree that the function runs well.

PERANCANGAN E-LEARNING BERBASIS WEB DI SMK SIDRATUL MUNTHAHA YAPIS WAMENA PAPUA DENGAN METODE BLACKBOX

Syarifah Syarifah, Kurniadin Abd Latif — Wed, 17 Jul 2024 08:08:32 +0700

In the era of technology which is developing rapidly in all fields, the learning process in the 3T Region, especially Sidratul Munthaha Yapis Wamena Papua Vocational School, still carries out a conventional learning process. This means that some students and teachers are not optimal in recognizing and implementing technology in the field of education. E-learning has developed and been applied in various schools. However, this convenience is not yet fully felt by students and teachers in Wamena Papua. The aim of this research is to design web-based e-learning and test the service level of the e-learning system using the black box method. The design method used is the waterfall method with stages of analysis, design, coding and testing. Testing was carried out using the black box method. The research results show that there are five interface displays for e-learning that have been designed, consisting of the homepage display, login page, classroom creation form, material posting form, and attendance display. Based on the results of black box testing, it proves that the e-learning system can be used according to the expected function. However, there are several notes on the design of this e-learning from media experts, including: the need to provide validation of assignment descriptions, automate assignment publication times, provide information on assignment collection pages, and automate assignment upload times.

IMPLEMENTASI LANDBOT SEBAGAI PENGHUBUNG DIALOGFLOW NLP DENGAN WEB INFORMASI KEPADA CIVITAS AKADEMIK

Fransiska Natalia, Albert Yakobus Chandra — Mon, 29 Jul 2024 13:41:38 +0700

Saat ini banyak universitas telah menggunakan sistem Informasi web untuk menyampaikan Informasi. Namun, masih sering terjadi kendala yang dialami mahasiswa dan calon mahasiswa yaitu sistem yang masih memberikan Informasi secara general, kesulitan dalam mencari informasi terkait perkuliahan, serta unit atau kontak admin yang diperlukan untuk menangani kendala yang dihadapi calon mahasiswa baru , baik di jam kerja dan di luar jam kerja.

Oleh karena itu, pada penelitian prototipe chatbot ini menggunakan sebuah platform berupa dialogflow dan landbot yang akan digunakan yaitu untuk mengatasi masalah yang dihadapi dan mempermudah para civitas akademik mendapatkan informasi. Adapun tahapan dalam pengembangan aplikasi chatbot ini yaitu analisis kebutuhan, desain, membuat prototipe, evaluasi dan rilis aplikasi. Metode yang digunakan dalam pembuatan chatbot ini yaitu Natural Language Processing (NLP).Proses pemberian pengetahuan dalam chatbot ini mencakup data terkait Informasi yang dibutuhkan sehingga apa yang ditanamkan dalam dialogflow dan landbot berguna untuk meningkatkan pemahaman konteks dan makna pertanyaan. Dalam hal ini pula, pelatihan model machine learning digunakan untuk meningkatkan keterampilan pemrosesan bahasa alami landbot. Berdasarkan hasil pengujian yang dilakukan aplikasi ini berjalan 100% dengan baik. Dalam hal ini menunjukan bahwa, chatbot ini mampu menjawab pertanyaan-pertanyaan yang diberikan, serta dapat mengenali pengetahuan yang telah diberikan sebelumnya.

OPTIMASI PENENTUAN LOKASI INDUSTRI DENGAN GIS-MCA: INTEGRASI TEKNOLOGI UNTUK ANALISIS SPASIAL MENDALAM

Ahmad Amri Nur, Dian Sandri, Naufal Haidar Ahmada, Revi Aulia Purbandini — Wed, 10 Jul 2024 00:00:00 +0700

and extinction if its location does not match its purpose. Each region requires an Industrial Designation Area for economic development, activity separation, environmental management and infrastructure development. The purpose of this study is to identify and find the location of the Industrial Designation Area in Purbalingga Regency that meets the technical criteria totaling seventeen criteria. These criteria are weighted through the Analysis Hierarchy Process technique to obtain the level of importance of all criteria by expert opinion. The method used is Spatial Multicriteria Analysis using the ArcGIS map processor. The results of the study show that the location of the Industrial Designation Area that is very suitable in Purbalingga Regency is 2,165 Ha spread across Bukateja, Kemangkon, Purbalingga and Kalimanah Districts. In-depth spatial analysis shows that the Very Suitable location is supported by the existence of a planned toll road, station, airport and is still close to the city center. The seventeen criteria used produce three categories of industrial area suitability levels, namely Very Suitable covering an area of 2,165 Ha, Suitable covering an area of 2,123 Ha and Quite Suitable covering 8,071 Ha.