Journal of System and Computer Engineering (JSCE)

Analisis Evaluasi Website Kalam UMI Berbasis Framework ISO/IEC 25010 Dengan Metode Goal Question Metric

2024-01-23T15:26:16+00:00

In this digital era, people cannot be separated from the unlimited need for information and communication thanks to the internet. One of the learning systems that utilizes technology is the Learning Management System (LMS). LMS is a technology system to manage and support the learning process, distribute lecture materials, and enable collaboration between lecturers and students. Universitas Muslim Indonesia (UMI) has an LMS called Kalam. The use of a system that is very intensive used by lecturers and students causes the need for evaluation to improve the quality of the learning process. ISO/IEC 25010 has become a standard recommendation because it has all the characteristics and sub-characteristics required for information system quality compared to other models. In this study, researchers used the Goal Question Metrics (GQM) method according to the characteristics based on the ISO/IEC 25010 framework in measuring the quality of the Kalam UMI website software. The evaluation of the Kalam UMI website on several characteristics has been included in the good and very good categories, but in security characteristics, there are sub-characteristics that still need to be improved, namely integrity with a percentage of 66.67%.

The quality sub-characteristics that have been categorized as good and very good are completeness of functionality with a percentage of 89.28%, correctness of functionality with a percentage of 93.75%, suitability of functionality with a percentage of 75%, confidentiality with a percentage of 75%, non-repudiation with a percentage of 100%, accountability with a percentage of 100%, authenticity with a percentage of 100%, effectiveness with a percentage of 82.96%, efficiency with a percentage of 83.06%, usability with a percentage of 84.47%, Trust with a percentage of 81.67%, pleasure with a percentage of 81.67%, comfort with a percentage of 86.46%, economic risk mitigation with a percentage of 88.75%, health and comfort mitigation with a percentage of 85.21%, environmental risk mitigation with a percentage of 90%, and flexibility with a percentage of 83.54%.

Optimalisasi Throughput Pada Penerapan Load Balancing Dalam Jaringan Cloud Menggunakan Round Robin dan Least Connection

2024-01-23T15:28:04+00:00

As the complexity of cloud infrastructure increases, one of the main challenges faced is how to optimize throughput in cloud networks. Optimal throughput is crucial to guarantee the performance of applications and services hosted in the cloud. Using load balancing techniques can optimize throughput, but careful analysis and testing is needed regarding the appropriate load balancing approach to improve throughput performance. This research aims to analyze the effectiveness of load balancing strategies with Round Robin and Least Connection in increasing throughput in cloud networks. Through empirical testing, this study compares the two methods based on the resulting throughput at various connection levels, ranging from 1000/600 to 5000/1000. Test results show that both strategies provide similar performance at low connection rates, with Round Robin recording slightly higher throughput at 1000/600. However, the difference becomes almost imperceptible as the number of connections increases. At 2000/700 to 5000/1000, the throughput generated by both strategies shows marginal differences, with Least Connection slightly superior at the highest connection level. These findings indicate that both Round Robin and Least Connection can be applied effectively to increase throughput in cloud environments, with the selection of specific strategies depending on workload characteristics and service requests. This research provides valuable insights for academics and IT practitioners who want to optimize cloud networks to achieve maximum performance

Implementasi Data Mining Menggunakan Algoritma Naïve Bayes Untuk Prediksi Kelulusan Siswa

2024-01-23T15:29:36+00:00

The student graduation process involves a series of stages and requirements that must be met by each student. Graduation is an important step in education because it determines the student's ability to continue to the next level. In this research, student graduation data is processed using data mining techniques, especially the Naïve Bayes method, to predict student graduation. Attributes such as Practice Scores, School Exams (US), National Exams (UN), and Student Behavior are used in this prediction process. The research results show that from a population of 60 students, the model successfully predicted that 45 students would pass, and 15 students would not pass. The Naïve Bayes method was implemented using the orange application and produced an accuracy level of 98.33% with a precision level of 100.00%. These results show that the Naïve Bayes method is very effective in predicting student graduation with a high level of accuracy. Therefore, this data mining technique can be a valuable tool for educational institutions in improving the quality of student graduation outcomes.

Mengungkap Opini Publik: Pendekatan BERT-based-caused untuk Analisis Sentimen pada Komentar Film

2024-01-23T15:30:54+00:00

BERT (Bidirectional Encoder Representations from Transformers) adalah model bahasa yang revolusioner dalam pemrosesan bahasa alami, mengandalkan encoder untuk menghasilkan representasi kontekstual dari teks input. Melalui pendekatan tokenisasi, embedding, dan attention mechanisms pada setiap layer transformer, BERT memungkinkan pemahaman hubungan antar kata secara mendalam dan bidireksional. Keunikan BERT terletak pada kemampuannya untuk memproses konteks dari kedua arah, menciptakan representasi vektor yang kaya makna. Model ini telah menjadi pionir dalam transfer learning di NLP, memungkinkan pemanfaatan representasi umum pada tugas-tugas khusus setelah proses pelatihan. Dengan demikian, BERT mengubah paradigma pemrosesan bahasa alami, membuka pintu untuk aplikasi yang lebih canggih seperti klasifikasi teks, analisis sentimen, dan pemahaman bahasa yang lebih kontekstual.

Aplikasi Berbasis Website Untuk Mendeteksi Status Gizi Balita Menggunakan Metode K-Nearest Neighbors (KNN)

2024-01-23T15:31:32+00:00

Detecting the nutritional status of toddlers is important in monitoring and caring for toddlers' health. In this study, we propose the use of the K-Nearest Neighbors (KNN) method to detect the nutritional status of toddlers based on relevant attributes such as weight, height, age, gender and nutritional intake. This research involves developing a computer-based application that uses the KNN algorithm to classify the nutritional status of toddlers. Toddler data that has been collected from legitimate sources is used to train the KNN model. After training, this model can predict the nutritional status of new toddlers based on the entered attributes. In our experiments, we test and evaluate the performance of KNN models using evaluation metrics such as accuracy, precision, recall, and F1-score. In practical applications, the KNN model can be used as an aid in determining the nutritional status of toddlers and providing recommendations for appropriate action, such as increasing nutritional intake or necessary medical care. This research makes a contribution to the field of monitoring and caring for toddler nutrition by combining the KNN method as a tool for detecting nutritional status. The application developed can help medical personnel and parents monitor and take appropriate action related to toddler nutrition. The application, which has been built in the form of a website, can help detect the nutritional status of toddlers. When applying the KNN method to toddler nutritional status data, the application was successful in detecting toddler nutritional status with an accuracy of 74.73%.

Sistem Pengendali Alat Elektronik Rumah Tangga Berbasis Internet of Things (IoT) Menggunakan Voice Recognition

2024-01-23T15:33:14+00:00

Using switches on household electronic devices that use electrical energy requires human action to get to the switch location and requires more time and energy. If human conditions do not allow for physical activity, such as for people with disabilities or people who are tired or sick, this is very difficult to do. The aim of this research is to create a system for controlling the use of household electronic devices based on the Internet of Things (IoT) using Voice recognition. This research was conducted at the Malone Celuler Shop in Bulukumba Regency. This type of research is Research and Development (R&D) with 6 stages, namely literature study and problem formulation, problem analysis, data collection, system design, system design implementation and system testing. The instruments used in this research were a laptop, Arduino Uno, PZEM-004T sensor, blynk application, voice recognition. The research results show that the system for controlling the use of household electronic equipment can turn it off and on effectively through voice recognition and provide information on whether household electronic equipment is still on when outside the home with the blynk application

Klasifikasi Ulasan Konsumen Menggunakan Random Forest dan SMOTE

2024-01-25T08:28:39+00:00

Ulasan berupa deskripsi teks direpresentasikan dalam bentuk skala peringkat tidaklah cukup sebagai acuan dalam menentukan sentimen yang seringkali bias. Penelitian ini dilakukan untuk mengklasifikasi ulasan konsumen dengan menerapkan teknik klasifikasi sentimen pada tingkat dokumen. Penelitian ini berfokus pada imbalanced class klasifikasi ulasan konsumen. Metode yang digunakan dalam mengatasi permasalahan imbalanced class adalah Synthetic Minority Oversampling Technique (SMOTE) dan Random Under-sampling (RUS) pada tahap pre-processing dan pada klasifikasi menggunakan Random Forest. Fitur data yang digunakan adalah fitur teks ulasan (proses analisis dan klasifkasi sentimen) dan fitur peringkat (proses pelabelan). Penerapan teknik class imbalanced (dalam hal ini SMOTE dan RUS) pada tahapan pre-processing mampu memberikan perubahan peningkatan akurasi dan mengenali data yang awalnya dianggap minor.

Rancang Bangun Sistem Cerdas Pendeteksi Kerusakan Mesin Pada Sepeda Motor 4 Tak

2024-01-23T15:35:25+00:00

The influence of using computer technology in solving a problem is very large, for example in determining motor damage. In this case, it is devoted to 4-stroke motorcycle vehicles. Currently, the average motorcyclist does not know about the symptoms or signs of damage to their vehicle, so motorists ignore these symptoms and consider these symptoms to be trivial problems, which will not cause damage to their motorcycles. Motorists will be aware of these symptoms after they can no longer use their vehicles. This research proposes to design an intelligent system that can help motorized vehicle drivers to find out earlier about symptoms or signs of damage to their vehicles and take appropriate action before more serious damage occurs to their vehicles. The results obtained are that an intelligent system can detect damage to a 4-stroke motor quickly and can facilitate the provision of solutions and the diagnostic process without a 4-stroke motor expert

Analisis Sentimen Ulasan Pengguna Tiktok pada Google Play Store Berbasis TF-IDF dan Support Vector Machine

2024-01-23T15:36:34+00:00

The rapid development of information and communication technology, particularly in social media platforms, has created an environment where users actively share their experiences and opinions related to various services and applications. One platform that has gained significant popularity is TikTok, a video-sharing application that has become a global phenomenon. With the increasing number of TikTok users, user reviews on distribution platforms such as the Google Play Store have become a crucial source of information. Sentiment analysis of these reviews can provide deep insights into how users respond to the application, while also offering valuable feedback for developers. The research aims to conduct sentiment analysis of TikTok user reviews on the Google Play Store using the Term Frequency-Inverse Document Frequency (TF-IDF) weighting method and the Support Vector Machine (SVM) algorithm as a classification method to achieve optimal results. There are three main stages: the initial stage involves data collection and data pre-processing, followed by the pattern recognition stage, which includes TF-IDF weighting and SVM classification. The final stage consists of evaluation and analysis. The opinion classification obtained includes three categories: positive, negative, and neutral. Based on the evaluation results, the proposed method successfully achieved high accuracy for the 70-30% training-testing split, reaching 84%. The conclusion drawn from these evaluation results indicates that the proposed method can be utilized in the sentiment analysis process of TikTok user reviews.

Klasifikasi Omset ATK menggunakan Algoritma Naïve Bayes

2024-01-23T15:38:12+00:00

Small businesses play a role in absorbing work energy, as a source of innovation, providing economic services to the community and in the process of equalizing and increasing community profit. One of the factors that influences the dynamics of modern business is technological progress. Evolving technology has opened the door to rapid innovation. Profit classification is a popular and effective approach for businesses that can tailor marketing strategies to each group more effectively which will impact business profit. The stock condition of an item greatly influences sales revenue. The increasing demand for goods will result in large profit. Product availability to meet consumer needs is a problem that must be overcome. The stock condition of an item greatly influences sales profit. This research is related to the classification of profit for each item sold in a shop, whether it is sold a lot or not enough to maximize the stock of goods per time using the Naive Bayes algorithm. From data research, attribute grouping, preprocessing, data transformation and modeling were carried out using the Naïve Bayes algorithm in the Python programming language. Testing the Naïve Bayes algorithm obtained 90% accuracy results for the classification of stationery profit. The system can determine the classes of goods that are sold a lot and goods that are not sold enough, becoming a solution for shop leaders to more easily take business strategies quickly and optimally which of course will affect profit.

Perancangan Sistem Monitoring Kualitas Udara Ruangan Berbasis Internet of Things (IoT)

2024-01-23T15:39:55+00:00

The purpose of the research is to monitor air quality in the room by using MQ2, Microcontroller NodeMCU equipped with ESP8266 wireless module by using MQTT protocol for Data Communication sensor node to server. System testing uses a black box testing approach.The type of research used in this study is experimental research with the approach of PPDIOO methodology. This research is done by conducting trials where mechanical and electronic design for hardware components designed to build tools using this sensor can work according to the objectives and target desired.The results of this research show that when the sensor detects the existence 150 of smoke, gas, carbon monoxide (CO) then the value of the ADC sensor will give a warning notification in the form of an alarm generated from the tone/tone of the alarm will be active when the sensor is actively reading the value of ADC with a tolerance above 150 ppm

UI/UX Design for Public Facility Complaint Services Using Mobile-Based Design Thinking Methodology

2024-01-23T15:40:56+00:00

This research examines the UI/UX design for public facility complaint services in South Tangerang City, using the Design Thinking methodology. The primary focus is to meet the needs of users, including complaint submitters and government officials. The prototype created was tested through Maze to collect data related to the application's usage. During testing, it was revealed that users encountered difficulties in completing specific tasks, especially in providing feedback on resolved complaints. There was also a high error rate in the login and complaint submission processes. Although government officials generally succeeded in completing their tasks, there were significant errors in the process. These findings indicate the need for further improvement, particularly in aspects not successfully addressed by users and the enhancement of efficiency in the complaint submission process. The importance of iteration and application design improvement is emphasized to enhance the overall user experience. These improvements are expected to more effectively meet user needs and increase satisfaction with the complaint service.

Sistem Pendukung Keputusan Penerima Kartu Indonesia Sehat Menggunakan Metode Analitycal Hierarcy Process

2024-01-23T15:41:37+00:00

Abstract

The Healthy Indonesia Card (KIS) is an identity card for National Health Insurance participants which is managed by the Health Social Security Administering Agency (BPJS). KIS is a program whose members are taken from underprivileged communities and the provision of cards is determined by the government and payment of contributions is borne by the government. However, sometimes in society there are many discrepancies in who is entitled to receive assistance and who is not entitled to receive assistance because the selection is still manual and based on recommendations from the government. This is the background that gave rise to the idea of creating a decision support system that could serve as a recommendation for selecting people who are entitled to receive KIS. The research design used is a Data Flow Diagram (DFD). The method used to carry out the selection process and obtain results is the Analytical Hierarchy Process (AHP) method. The results of the design provide calculation results for the ranking of those entitled to receive KIS, providing reports on people who are and are not entitled to receive KIS.

Analisis Kekuatan Pencahayaan Lampu Dalam Sistem Penerangan Jalan Umum Tenaga Surya 200wp

2024-02-15T01:59:00+00:00

Lighting is a vital requirement for elements of human activity, especially during productive hours of human activity. The important role of electricity in supporting the progress of lighting process technology, mechanical and other automation cannot be missed. Light is an electromagnetic wave consisting of a magnetic field and an electric field that are perpendicular to each other and light can also experience reflection, diffraction, refraction, polarization and interference. Light in its design is a lumen with flux units emitted in solid angle units by a source with a light intensity of one Candela. Lux is a metric unit of measurement for light on a surface. Efficacy is the average illumination / brightness achieved in a flat work area per watt in general lighting in the room expressed in lux/Watt/m2. Field studies, namely by observing and collecting data directly in the field for research objects on 40 watt LED lights installed in Solar Public Street Lighting on the Krisnadwipayana University Campus, as a reference for the design and installation and location of Solar Public Street Lighting placement. Testing the lighting data collection point (lux) is carried out to determine the angle of emitted lighting (lux) and the level of lighting radius obtained based on field data. As a result to find out according to standardization (SNI 7391; 2008). The street lighting for theenormous Solar Street in the parking lot of the Krisnadwipayana University Laboratory uses LED lamps with a flux of 5778 lumens and a power of 40 watts. The average lumen from the results of observations of data collection corresponds to the height distance between the LED 6 meters and the lux meter, and the angle of 5o is 53 lux.

Keyword : Light, Lumen, Lux and Efficacy, Electrical Energy.