Sinkron : jurnal dan penelitian teknik informatika
Not a member yet
1261 research outputs found
Sort by
Improving Tesseract OCR Accuracy Using SymSpell Algorithm on Passport Data
No full textOptical Character Recognition (OCR) is a technology used to recognize text from images or digital documents, such as passports. One popular OCR tool is Tesseract as it offers high accuracy. However, OCR accuracy is often affected by various factors, including image noise and/or non-text elements. This article discusses the application of the SymSpell algorithm for post processing to improve OCR accuracy on standard Indonesian passports. OCR will be focused on the Visual Inspection Zone, specifically the Place of Birth and Issuing Office values. Unlike the Machine Readable Zone which is composed of individual codes and a clear background, the Visual Inspection Zone often experiences OCR errors due to holograms blocking the text and spaced layouts. SymSpell is an edit distance based spelling correction algorithm designed to process data quickly and efficiently, even on very huge datasets. In this study, SymSpell is used to detect and correct errors in OCR results that are compared to a corpus word list. Experimental results with 10 tested scans and passport photos showed that the integration of SymSpell with the Research and Development methodology was able to improve the OCR accuracy rate by 21,43% for certain Place of Birth and Issuing Office data from the Visual Inspection Zone. With this approach, OCR systems can provide more reliable results for practical applications
Development of Augmented Reality-Based Learning Media for Solid Geometry for Elementary School Students
No full textAR-based media is expected to create a more interactive and engaging learning experience, enhance students’ understanding, and motivate them to learn independently and actively in the digital era. The data collection stages applied include testing, observations, and the distribution of questionnaires. The development of this learning media follows the MDLC model, which aims to design improvements to the existing system. The results of the blackbox testing, the "Bangun Ruang" application is proven to be valid and successfully used, with excellent results in the SUS test, where the Ease of Use score reached 86%, Efficiency 88%, Effectiveness 90%, and Satisfaction 87%. This indicates that the application has high levels of usability, efficiency, and effectiveness, while also providing a satisfying user experience. The application not only operates according to the designed specifications but also provides a positive user experience. Thus, it can be widely used, especially in educational environments, to help students understand geometric concepts in a more interactive and engaging way. Developers can continue maintaining and improving the application based on user feedback to ensure it remains optimal and aligned with users' needs in the future
Sentiment Analysis Using Grok AI as an Auto-Labeling Tool in The Text Processing Stage
No full textA critical aspect of Natural Language Processing (NLP) is text processing, where text labeling represents the most significant challenge due to its resource-intensive nature when conducted manually. At this stage, automatic labeling emerges as a more practical solution, particularly with the advent of Artificial Intelligence (AI), which offers tools to address this obstacle. Grok AI, equipped with a new feature operable on Platform X, provides a promising approach. This study aims to leverage the Grok AI feature on Platform X for automatic text labeling. The research methodology involves labeling text data obtained from a public dataset. To assess the quality of the labeling results, an evaluation method employing Naive Bayes classification modeling is applied. The findings reveal that Grok AI's performance closely approximates that of human labeling. The highest accuracy achieved by Grok AI is 51.71% using the k-Nearest Neighbors (k-NN) algorithm, approaching the human labeling accuracy of 60.52% with k-NN. Furthermore, Grok AI surpasses the performance of VADER labeling, which achieves an accuracy of only 49.49% with Naive Bayes. Consequently, the Grok AI feature on Platform X presents a viable alternative for the automatic labeling of text data
Evaluation of Mobile Academic Information System with Notifications Using Heuristic Evaluation and WCAG-EM
No full textSIsKA-NG Mobile is a mobile information system that provides specific information related to students’ study activities in the Computer Science Study Program at Universitas Pendidikan Ganesha. This newly developed application has never been evaluated. Therefore, it is necessary to evaluate the accessibility and interface for improvement. This study focuses on the analysis related to the application accessibility of partial color blindness users, using the WCAG-EM method and the heuristic evaluation method. Based on the accessibility evaluation results on the first development, it was found that SIsKA-NG mobile did not fully meet the standards of the mobile application, while based on the results of the heuristic evaluation, it was found 24 problems, where 8 problems with the highest severity rating (Catastrophic), 7 problems with high priority (Major), and others related to low priority (Minor) problems and insignificant problems. The subsequent development to enhance the quality of the interface and accessibility of SIsKA-NG Mobile used those findings as a reference so that this application can meet the needs of all users more effectively and inclusively
Food Recipe Recommendation System with Content-Based Filtering and Collaborative Filtering Methods
No full textCooking your own food at home is a good step toward reducing fast food consumption. Fast food increases the risk of dangerous diseases. The diversity of recipe information available on the internet makes it difficult to choose recipes that match user preferences. Mobile technology can help with this by recommending recipes that better suit users' eating habits. This makes the transition to a healthier diet easier. Therefore, in this study, a recommendation system was developed that can recommend recipes based on the preferences of Android users. Two main recommendation methods are used in this study: content-based filtering and collaborative filtering. Using cosine similarity, a content-based recommendation system identifies the proximity between a recipe for food and its related context. The history of user comments on recipes serves as implicit feedback for the collaborative recommendation algorithm. This eliminates the need for explicit evaluations, such as ratings. This recommendation system generates recommendations in the form of the top ten food recipes with an evaluation matrix, referred to as NDCG@k and Hit-Ratio@k. The tests revealed that a content-based filtering technique may produce helpful recommendations, with the highest similarity score of 0.41 for the entry "chocolate cake that you can easily make at home." Meanwhile, in the collaborative filtering method using the Neural Collaborative Filtering (NCF) approach, the system shows consistent performance improvements, with the MAP@10 value increasing from 0.705 to 0.767 and the NDCG@10 from 0.78 to 0.83 after 10 training epochs.
Keywords: Recommendation systems; content-based filtering; neural collaborative filtering; cosine similarity; implicit feedbac
Design of Real-Time Project Monitoring Dashboard Using Kimball’s Data Warehouse Approach and Google Data Studio
No full textThe growth of the construction industry in Indonesia triggers an increasing need for an efficient project management system, especially in presenting project data accurately and in real-time. PT Dream Island Development (PT DID), a specialist MEP contractor company, faces challenges in presenting project reports to executives because the data is still presented in the form of Excel tabulations which require up to three days of processing time and are difficult to interpret quickly. This research aims to design an interactive dashboard-based project data visualization system using Google Data Studio (Looker Studio) to present project information intuitively and responsively. The method used includes a software engineering approach with five main stages: requirements analysis, data warehouse design, ETL process using Pentaho Data Integration, visualization using Google Data Studio, and testing using User Acceptance Test (UAT). Project data from 2022-2024 was modeled using a star schema and displayed in four main dashboards: project cost, project value, project progress, and details per project. The test results showed a high level of user satisfaction with a functionality score of 93.5%, reliability 91.33%, usability 96%, and efficiency 94.66%. These findings indicate that the developed system effectively supports PT DID's needs in project monitoring and data-based decision-making. The system also has the potential to be replicated in other construction companies as an efficient and scalable business intelligence solution.
This research contributes to the growing body of construction informatics by integrating Kimball’s nine-step methodology with modern data visualization tools to enhance project transparency and decision-making
A Multi-Objective Decomposition Model for Integrated Urban Transit Line Planning and Passenger Routing
No full text: Urban public transport networks must balance traveler convenience with tight budgetary and capacity constraints. This study develops a comprehensive multi-objective integer programming framework that unifies line selection, frequency setting, and passenger routing to minimize door-to-door travel time and operating cost while respecting vehicle capacities and limiting transfers. The model is solved using a Dantzig–Wolfe decomposition approach with linear-programming relaxation, which enables tractable solutions on realistically scaled networks. To reflect real-world commuting behavior, three increasingly sophisticated formulations are proposed: a Basic Line Planning Model, a Direct Connection Capacity Model, and a Change-and-Go Model that embeds walking and waiting penalties. On a six-edge, four-node network with 6,000 passenger trips, the Change-and-Go Model emerges as the most effective, reducing average travel time by 47%, halving transfers, and increasing cost by only 11% compared to the incumbent plan. Sensitivity analysis reveals that the model remains robust under varying demand levels and cost–time priorities. The proposed framework thus offers a scalable and passenger-friendly decision-support tool that significantly improves public transport efficiency with moderate investment, making it especially valuable for urban transit agencies seeking to modernize their services
Enhanching Prophet Time Series Forecasting on Sparse Data via Hyperparameter Optimizattion: A Case Study in Retail
No full textIn today’s competitive business landscape, accurate sales forecasting is crucial for retailers to optimize inventory, prevent overstock, and support strategic decision-making. However, many small to medium enterprises operate with sparse and irregular sales data, making conventional forecasting methods less effective. This study aims to evaluate the performance of the Prophet time series model in such non-ideal conditions and to investigate how hyperparameter tuning affects its forecasting accuracy. The research adopts the Prophet algorithm, an additive time series forecasting model developed by Facebook, which incorporates trend, seasonality, and holiday components. The model was implemented in two configurations: one using default parameters, and another with manually tuned hyperparameters, including changepoint prior scale (CP), seasonality prior scale (SP), and seasonality mode. A total of 32 experiments were conducted using historical transaction data from PT Eko Hejo. Results show that the default Prophet model achieved a MAPE of 9.50%, while the best-performing configuration (CP = 0.5, SP = 0.01, additive mode) reduced the MAPE to 6.80%. This indicates that hyperparameter tuning significantly improves forecast accuracy, even in sparse data environments. The study contributes both practically and scientifically by demonstrating that Prophet, when properly configured, is a robust and adaptable tool for business forecasting with limited data. It also highlights the value of manual tuning in enhancing model responsiveness and generalization, offering insights for further research in model comparison, automated optimization, and hybrid forecasting approaches
Clustering IT Incidents Using K-Means: Improving Incident Response Time in Service Management
No full textIncident management is one of the critical processes in Information Technology service management that aims to manage disruptions and minimize the impact of unexpected incidents on business services. This study applies the K-Means algorithm to cluster IT service incidents, aiming to enhance company operational efficiency. Utilizing a dataset from the UCI Machine Learning Repository comprising 141,712 events related to 24,918 incidents, this research analyzes incident patterns and characteristics for optimized handling. The data was analyzed through a series of preprocessing stages, and the elbow and silhouette methods were used to determine the optimal number of clusters. From the results, it was successfully grouped into 4 (four) clusters with a distortion score value of 964264294.569 and 0.52 silhouette score based on incident characteristics, such as urgency, priority, and number of reassignments. From this, the clustering results show that the K-Means algorithm effectively identifies incidents that require further handling, such as those with high urgency and priority, as well as helping the company focus resources to resolve incidents that have the most impact on the business sector. This research provides a data-driven solution to improve incident management and Service Level Agreement (SLA) fulfillment, while offering a framework for more effective and efficient IT incident analysis and resource allocation
Meeting Room Booking System with WhatsApp Notification Feature Using Extreme Programming Methods in RS Muhammadiyah Lamongan
No full textMeeting room management in hospitals plays an important role in supporting operational efficiency and coordination between departments. At RS Muhammadiyah Lamongan, common issues such as overlapping schedules, delays in booking information, and lack of transparency in the scheduling process are still frequently encountered. This study aims to develop a web-based meeting room booking system using the Extreme Programming (XP) method, integrated with a WhatsApp notification feature. The system is designed to improve transparency, minimize scheduling conflicts, and enhance communication between administrators and users. Requirements gathering was conducted through interviews with the hospital's secretariat, and the system was developed using the Laravel Framework and WhatsApp API. The system testing was carried out using Blackbox Testing and User Acceptance Testing (UAT) with a Likert scale. The test results showed that the system achieved a perfect score of 100 out of 100 points, indicating that all core features functioned as expected without significant technical issues. This system is expected to serve as an effective solution to support a more efficient, real-time, and structured meeting room scheduling process at RS Muhammadiyah Lamongan