JTIM : Jurnal Teknologi Informasi dan Multimedia
Not a member yet
290 research outputs found
Sort by
Studi Pemodelan dan Prediksi Aktivitas Antibakteri Biopo-limer Kitosan Menggunakan Response Surface Methodology (RSM)
No full textInfections occured in the human are mostly caused by uncontrolled growth of Staphylococcus aureus bacteria. A strategy to inhibit bacterial growth can use antibacterial agents such as chitosan. The mechanism of the effectiveness of chitosan as an antibacterial is quite complex, even the data on its antibacterial activity is quite fluctuating so that it is difficult to analyze accurately and efficiently. Therefore, the purpose of the study was to predict the inhibition zone of s.aureus bacteria through laboratory experiments combined with modeling using the Central Composite Design (CCD) approach. The research was carried out with two main stages, including chitosan isolation and calculation of bacterial inhibition zones. The production of chitosan leverages the microwave isolation and FTIR to examine for the degree of deacetylation and its functional group using. Furthermore, the antibacterial activity of chitosan biopolymer was tested using the diffusion method combined with modeling using the RSM CCD approach. The results showed that chitosam from oyster shell was obtained by DD of 83.29% and the emergence of typical chitosan groups, such as amine (NH2) and hydroxyl (OH). Chitosan can hamper the growth of s. aureus bacteria with an inhibition zone of up to 0.40 mm. The experimental data were combined with computational modeling obtained the values of the determination coefficient R2 = 0.6083. The modeling was assessed by p-value of < 0.0001 and F-value of 13.46. Statistically, the obtained model is relevant to the relationship between the number of bacterial colonies and the concentration of chitosan solution with the bacterial inhibition zone. Based on numerical analysis and modeling, the predicted values of the number of s. aureus bacterial colonies and chitosan concentrations were 550,000 CFU/ml and 42.5%. Therefore, Pearl shells can be isolated into chitosan, as well as chitosan has the potential to be a good antibacterial agent. The model has good prediction performance, but it rquires to increase the number of point spreads and it is necessary to validate the prediction results to obtain actual predictions
Pengembangan Media Pembelajaran Interaktif Gerakan Shalat Berbasis AR (Studi Kasus: Madrasah Diniyah Al-Barokah)
No full textFiqh learning in madrasah diniyah often still relies on conventional media such as textbooks, which tend to be less engaging for students and may lead to decreased interest. This study aims to develop an interactive learning media for shalat movements using Augmented Reality (AR) technology and the ADDIE development model (Analysis, Design, Development, Implementation, Evaluation). AR was chosen as an innovative solution to deliver fiqh material in a more visual, interactive, and contextual manner. The application was designed using a marker-based approach and visualizes eight shalat movements through 3D animated objects validated by Islamic scholars. The trial was conducted at Madrasah Diniyah Al-Barokah involving 8 students and 2 teachers. Black-box testing confirmed that all features functioned properly across different devices. Meanwhile, the User Acceptance Test (UAT) showed a very high level of user satisfaction, with an average score of 91.22%, particularly in terms of ease of use, visual appeal, and clarity of content. Unlike previous studies, this application specifically focuses on visualizing shalat movements for children in madrasah diniyah within the context of fiqh learning. This research recommends future development with broader user testing and the integration of quantitative learning outcome evaluations
Penerapan User Centered Design untuk Optimisasi User Experience Aplikasi Virtusee
No full textThis study aims to redesign Virtusee, an employee performance monitoring platform, using the User-Centered Design (UCD) methodology. Virtusee features functionalities such as leave requests, monthly performance tracking, and self-service payslip access. The application, last updated in 2014, was described by users as outdated and not aligned with current trends. Based on the System Usability Scale (SUS) questionnaire, the application scored 40, categorized as poor with a grade of F, indicating the need for significant improvements. Employing the UCD approach, this study prioritized user needs and preferences in the redesign process, following these stages: (1) defining the usage context, (2) identifying user and organizational requirements, (3) design and implementation, and (4) usability evaluation. Testing the new design involved users who had used Virtusee more than once in the past month. The testing scenarios included exploring the new design, completing SUS and QUIS (Questionnaire for User Interaction Satisfaction) surveys via Google Forms, and conducting brief interviews to gather suggestions and critiques. Data were collected through observation, interviews, literature review, and questionnaire dissemination. Results showed a significant improvement, with the SUS score increasing to 80.3, indicating acceptable usability. The QUIS evaluation revealed average scores ranging from 5.9 to 6.8 across various indicators, exceeding the expected median. These findings highlight that the UCD methodology is effective in designing applications that are more user-centered, enhancing productivity and user satisfaction. This study provides valuable insights for developers aiming to create applications that are not only functional but also adaptive to evolving user needs, serving as a reference for designing solutions that align with user-centric principles
Perbandingan Support Vector Machine, Random Forest Classifier, dan K-Nearest Neighbour dalam Pendeteksian Anomali pada Jaringan DDos
No full textA Distributed Denial of Service (DDoS) attack poses a serious threat to network security and can disrupt online services by overwhelming the target server with excessive traffic. Effective detection of DDoS attacks requires a system capable of identifying anomalies in network traffic. In this context, Machine Learning (ML) offers an effective approach for classification and anomaly detection. However, different ML algorithms have varying strengths and weaknesses when processing large and complex network data. Therefore, this study aims to evaluate the performance of three ML algorithms: Support Vector Machine (SVM), Random Forest Classifier (RFC), and K-Nearest Neighbors (KNN) in detecting DDoS anomalies. The dataset used consists of 225,745 data points with 85 attributes that describe various characteristics of network traffic, such as destination port, flow duration, packet count, and packet size. This dataset is classified into two classes, BENIGN and DDoS, representing normal traffic and DDoS attacks, respectively. Evaluation is performed using several performance metrics, including accuracy, precision, recall, MCC (Matthews Correlation Coefficient), F-Measure, ROC Area, PRC Area, True Positive Rate (TPR), and False Positive Rate (FPR). The results show that the Random Forest Classifier (RFC) delivers the best performance with an accuracy of 99.99%, precision of 99.98%, recall of 100%, and a very low FPR of 0.02%. This is followed by the Support Vector Machine (SVM) with an accuracy of 99.91%, and the K-Nearest Neighbor (KNN) with an accuracy of 99.98%. All three algorithms demonstrate strong performance in detecting DDoS anomalies, with RFC slightly outperforming others in terms of consistency and higher classification capability. The findings of this study provide valuable insights for selecting the best algorithm to detect DDoS attacks in networks
Rancang Bangun Aplikasi Presensi Dan Penggajian Karyawan Berbasis Mobile Pada CV. Maju Berkah Sosa Julu (Mabes-J)
No full textThe attendance and payroll system currently implemented at CV. Maju Berkah Sosa Julu (Ma-bes-J) still uses a manual method or is still done by signing the attendance book. Every employee who comes to work is required to record their arrival time on the daily attendance sheet. The cri-teria for employee payroll is attendance, so the payroll system must be stricter in entering em-ployee attendance criteria, with slow and manual data processing also resulting in delays in sal-ary payments and the potential for miscalculation. As a result, not only employees are disadvan-taged, but also the company that has to bear the burden of employee dissatisfaction. The purpose of this study is to build a system that can increase employee productivity and satisfaction at CV. MABES-J. This application is expected to provide convenience for employees in recording at-tendance and providing clear information about the salary they receive. Researchers use the Agile method to develop the system, emphasizing flexibility and adaptation to change. The Agile ap-proach allows iterative system development, with a focus on collaboration and quick response to feedback. This study produces a system that brings significant changes in employee attendance management at CV. Maju Berkah Sosa Julu (Mabes-J). Before this system, the attendance record-ing process was done manually, which was often time-consuming and prone to errors. With the implementation of this new system, employees no longer need to fill in attendance manually. The system automatically records all attendance data, and integrates it into a centralized database
Prediksi Gender Berdasarkan Nama Menggunakan Kombinasi Model IndoBERT, Convolutional Neural Network (CNN) dan Bidirectional Long Short-Term Memory (BiLSTM)
No full textThis study proposes a name-based gender prediction model in the Indonesian language by combining the architectures of Indonesian Bidirectional Encoder Representations from Transformers (IndoBERT), Convolutional Neural Network (CNN), and Bidirectional Long Short-Term Memory (BiLSTM). The non-standardized and diverse structure of Indonesian names presents a significant challenge for text-based gender classification tasks. To address this, a hybrid approach was developed to leverage the contextual representation power of IndoBERT, the local pattern extraction capability of CNN, and the sequential dependency modeling strength of BiLSTM. The dataset consists of 4,796 student names from Universitas Bumigora, collected between 2018 and 2023. The preprocessing steps include lowercasing, punctuation removal, label encoding, and train-test splitting. Evaluation results based on accuracy, precision, recall, and F1-score indicate that the IndoBERT-CNN-BiLSTM model achieved the best performance, with an accuracy of 90.94%, F1-score of 91.03%, and training stability without signs of overfitting. This model demonstrates high effectiveness in name-based gender classification and holds strong potential for applications such as population information systems, service personalization, and name-based demographic analysis
Pengujian Usability Testing Menggunakan Metode System Usability Scale untuk Menilai Aplikasi SIKUU
No full textSIKUU or sistem informasi kendali suhu dan lampu is an android based application created with the purpose to monitoring temperature, light density, number of people, and electrical power consumption in a room. SIKUU needs to be tested to determine the level of ease and the purpose of application. Usability testing is the method for testing it. An easy method to use for usability testing is the system usability scale (SUS). SUS Method very popular because of its advantages such as involves end user directly, no special criteria for selecting end user, can be used for fewer respondents but the results are still reliable, the test scale is easy to understand, and the results are effective because it can immediately categorize the application as usable or not. Then, for SIKUU, usability testing was carried out involving 35 respondents from Institute Technology of Sumatera particurally from various study programs. These respondents were introduced to the SIKUU for the first time. The final SUS score is 71,14 which indicates that the acceptability level is in the acceptable category, and adjective rating in the good category. Meanwhile, i terms of the SUS percentile rank, the score falls within category C. In this study, the primary focus of the usability testing was solely on the final score. Therefore, form the users\u27 perspective, SIKUU is considered feasible for use
Implementasi Sistem Pengadaan Material pada SAC dengan Metode Waterfall
No full textDecision support systems in the material procurement process are important solutions to improve operational efficiency and accuracy, especially in retail companies such as SAC (Store Adede Cikampek) which is engaged in the sale of dolls. This study aims to design and build a web-based material procurement system that is able to manage the ordering process, stock recording, verification of incoming goods, and reporting automatically. The system development was carried out using the Waterfall method because its systematic stages are very suitable for handling the material procurement process at SAC which was previously manual and undocumented. With the Waterfall approach, each stage such as needs analysis, design, implementation, testing, to maintenance can be carried out in a structured manner, thus ensuring that the system built is able to overcome problems such as late ordering and errors in recording raw materials. At the implementation stage, this system was developed with various features such as supplier data management, raw material stock management, order history, and periodic report generation. To ensure the effectiveness of the system, testing was carried out using the System Usability Scale (SUS) approach involving twenty respondents from internal operational parties. The evaluation results showed that the developed system succeeded in meeting user needs and increasing the effectiveness of the procurement process by obtaining an average score of 96 which was categorized as "Excellent". This system is also considered easy to use, efficient, and can support the decision-making process in real time. It is expected that the implementation of this system can not only solve the problem of material procurement in SAC, but can also be used as a model for implementing similar systems in similar businesses. This research provides a practical contribution in the development of an integrated information system to support more optimal business processes
Acoustic Analysis on Cleft Lip Speech Signal
No full textCleft conditions significantly disrupt phonetic articulation, leading to hypernasality and irregular resonance characteristics. In this study, the formant analysis of normal and cleft speech is presented, with the aim of investigating acoustic differences in formant frequencies between cleft and normal speech using real-word utterances, focusing on the articulation of plosive consonants and resonance variability. The dataset consisted of 280 speech signals (140 cleft and 140 normal) uttering word /paku/. The speech signals were resampled to 16kHz and the silence in the speech was removed, next stage was followed by extracting the first three formants using the Burg algorithm. Statistical analysis revealed that the value of F1 and F2 in cleft speech were higher, alongside greater variability in formant distribution. Further analysis of plosive articulation highlighted irregular formant transition in cleft speech, indicating compromised intraoral pressure control. Additionally, a moderate negative correlation (r = -0.423, p<0.001) between F1 and F3 suggests a spectral pattern indicative of hypernasality. This finding underscores the potential of formant-based acoustic features as objective markers for early clinical assessment and provides a foundation for the development of diagnostic models in cleft speech research
Analisis Prediksi Penjualan Isi Ulang Air Galon menggunakan Metode LSTM dan SARIMA
No full textRefillable drinking water depots often face challenges in dealing with unpredictable customer demand on a daily basis. This uncertainty complicates the process of stock management, production planning, and overall operations. Without accurate sales forecasts, depots risk losing potential sales and experiencing a decline in service quality to customers. Therefore, a solution is needed that can accurately predict daily sales. The first step in this research is to collect relevant data. Once the data is available, pre-processing is conducted to prepare the data before entering the modeling process. The Long Short-Term Memory (LSTM) model has the advantage of remembering historical patterns in time series data. Meanwhile, the Seasonal Autoregressive Integrated Moving Average (SARIMA) model is an extension of ARIMA that can handle data with seasonal characteristics. In this study, the LSTM model demonstrated better performance than SARIMA. This is evidenced by the performance evaluation values: MAPE of 9.54%, RMSE of 0.17, and MAE of 0.14 for the LSTM model, which are lower than MAPE of 10.51%, RMSE of 0.19, and MAE of 0.16 for SARIMA. These values indicate that LSTM is capable of providing more accurate prediction results. Based on these results, it can be concluded that the LSTM model is more effective and recommended for use in predicting daily sales of refillable water at the Manshurin Water depo