JSAI (Journal Scientific and Applied Informatics)

Analysis of AI-Based Software Development (Gemini) Using ACM Code of Ethics Implementation

Rexcy Elsan — 2025-06-01

This research aims to analyze the application of the ACM Code of Ethics in the development of AI Gemini software by Google DeepMind. With the development of artificial intelligence (AI) technology, developers need to pay attention to ethical aspects in every stage of system development. This research uses a qualitative approach with a data triangulation method, which includes document analysis, interviews with AI experts, and a literature study. The results show that although Gemini AI has complied with some key ethical principles in the ACM Code of Ethics, such as data security and privacy (90%) and human well-being (85%), there are still challenges related to transparency (60%) and bias mitigation (70%). Lack of transparency in algorithms and training data as well as potential bias in results generated by AI are the main issues that need to be improved. Based on the results, overall the Gemini AI has shown fairly good ethical compliance with an average of 76.67% but still requires improvement in the aspects of transparency, bias mitigation, and usage regulation.

Optimizing Network Management with Bandwidth Limitation and Blocking Certain Access at PT Laxo Global Access by Applying the NDLC Method

Adiffanani Ramdansyah — 2025-06-01

This research focuses on optimizing network management at PT Laxo Global Akses by implementing bandwidth restrictions and restricting access to non-work related websites and applications. Internet usage in the company was previously unrestricted, resulting in decreased productivity and unstable network performance, especially during peak hours. To address these issues, this research applies the Network Development Life Cycle (NDLC) methodology and utilizes MikroTik devices for configuration. Key features such as Queue Tree for bandwidth distribution, Firewall Filtering for access control, and Hotspot User Management for user authentication were implemented. The redesigned network topology allows for more structured traffic management and real-time monitoring. The test results show that all test scenarios have succeeded as expected, so it can be concluded that the network configuration and policies implemented in the NDLC process were 100% successful at the implementation and testing stages.

Analysis of Maturity Alignment of Business Strategy with Information Technology at SAMSAT Bengkulu City Using Strategic Alignment Maturity Model (SAMM)

Kurnia Anggriani — 2025-06-01

Motor vehicle tax payments represent a substantial portion of regional income. However, recurrent delays in motorbike tax payments, notably in the Kandang Limun sub-district of Bengkulu City, pose a significant threat to regional development. This research aims to address these delays by predicting and implementing effective strategies. Using the RapidMiner application and the Artificial Neural Network (ANN) method, a delay rate of 36.14% was predicted. The study incorporates interviews to propose several strategies, including enhancing websites, improving offline and online services, implementing Drivethru services, and using SMS Blast notifications to combat delays. To evaluate the feasibility of these strategies, the research employs the Strategy Alignment Maturity Model (SAMM). The assessment at BPKD reveals a maturity level of 4, signifying an ability to improve and manage processes effectively.The maturity assessment suggests that the proposed strategies for overcoming delays are well-aligned with BPKD's capabilities. They have demonstrated a robust alignment between IT and business strategies, leveraging IT to enhance performance, and conducting thorough business and IT evaluations.In addition to the maturity assessment, the research offers suggestions and evaluation strategies. These encompass the adoption of biometrics for authentication, implementation of chatbots, Samsat self-service kiosks, real-time vehicle monitoring, and the introduction of automatic tax programs.

Implementation of NLP Election News Classification Using LSTM Algorithm

Harry Vadilan Sianturi — 2025-06-03

This study examined the application of a Long Short-Term Memory (LSTM)-based text classification method to categorize election news according to presidential and vice-presidential candidate entities. The core problem addressed was the lack of an automated classification system capable of identifying political affiliations directly within the vast volume of digital news content. In this research, news data were collected from open-access sources and automatically labeled based on the occurrence of candidate-related keywords. A supervised learning approach was implemented using the LSTM architecture to capture sequential patterns within the news text. The evaluation results demonstrated that the model achieved a validation accuracy of 95.44% and a macro-averaged F1-score of 0.95, indicating strong classification performance across all candidate categories. Furthermore, predictions on test data revealed the model’s consistency and stability in recognizing political entities. This study confirmed the effectiveness of the LSTM-based approach for entity-based election news classification and highlighted its potential for integration into automated media analytics and political discourse monitoring systems.

Implementation of Agile Scrum Method and Return On Investment (ROI) Analysis on Library Information System

Muhammad Hafiz Ar Rahman — 2025-06-03

The library is one of the important institutions that provide and manage information to support learning, teaching, and research activities. In its development, the need for an integrated and efficient library information system is increasing, along with the demands of digitizing services and managing data quickly and accurately. This research aims to implement the Agile Scrum method in the development of Library Information Systems and conduct a return-on-investment (ROI) analysis to measure the effectiveness of the investment from a business and economic perspective. Based on the research problem raised and the results of the research that has been conducted, it shows that the Scrum method is proven to be effective in increasing flexibility and responsiveness to user needs. Key features are completed on time, and user feedback can be responded to in the next sprint. Active collaboration minimizes the risk of errors and maintains system quality. The first-year ROI is positive although small (4.65%), but the system is sustainable, so the ROI will increase significantly in years 2 and 3 because costs remain low while benefits continue to grow.

Analysis of Archer.exe Malware to Identify Potential Threats on Operating Systems Using the Hybrid Analysis Method

Juan Haniful Kahfi — 2025-06-03

Cybersecurity has become a primary concern in the digital era, particularly regarding malware attacks targeting the Windows operating system. This study aims to analyze a Remote Access Trojan (RAT)-type malware named archer.exe, obtained from the Any.run platform. The method used is hybrid analysis, a combination of static analysis and dynamic analysis, to provide a comprehensive understanding of the malware's structure and behavior. Static analysis results show that archer.exe is a Portable Executable (PE) file with a size of 829.35 KB and employs packing techniques to conceal its payload. Meanwhile, dynamic analysis reveals that the malware modifies system registry keys, spawns child processes such as rundll32.exe and cmd.exe, and establishes a network connection to a Command and Control (C2) server at IP address 192.169.69.26 via the domain dominoduck2101.duckdns.org. These findings indicate that archer.exe poses a high risk of remote system access, data theft, and malicious background activity without user awareness. This study demonstrates that the hybrid analysis method is effective in identifying hidden threats and malicious behavior of RAT-type malware on Windows 10 systems.

Analysis of Canva and Gamma User Satisfaction Among Students Using the System Usability Scale (SUS) Method

Helen Anggraini — 2025-06-03

Canva and Gamma are web-based graphic design platforms widely used by students to create brochure, presentations, and social media content in a simple and efficient way. The purpose of this study is to evaluate user satisfaction. with both applications using the System Usability Scale (SUS) method. The study involved 100 students from Universitas Muhammadiyah Bengkulu who had used both platforms. The results indicate that Canva achieved a SUS score of 80.2, categorized as Excellent, while Gamma received an average score of 72.4, categorized as OK. These findings suggest that although both applications demonstrate good usability, Canva is perceived to be superior in terms of ease of use, feature integration, and learning curve. The outcomes of this study can serve as a reference for’s developers with educational institutions on selecting or developing design tools that are more responsive to user needs.

Designing a Web-based Teacher Attendance System Application Using Geo Fencing at XYZ Elementary School

Faizul Anwar Ramdhani — 2025-06-08

The web-based teacher attendance system application with Geo-Fencing technology integration is designed to improve the accuracy and efficiency of the teacher attendance recording process in the school environment. The use of Geo-Fencing allows attendance to only be done when the user is in a predetermined area, thus minimizing the potential for fraud and data manipulation. This research aims to develop a web-based attendance application and evaluate its usability level using the System Usability Scale (SUS) method. The evaluation was conducted on 20 respondents consisting of teachers and school administrators. Based on the test results, an average SUS score of 75 was obtained, which is included in the Good usability category. Thus, this application is considered quite easy to use, effective, and acceptable to users. The results of this study indicate that the web-based teacher attendance system application with Geo-Fencing has the potential to be widely implemented in the school environment, with some further development recommendations to improve user experience.

Analysis of Canva and Slidego User Satisfaction Among Students Using the Customer Satisfaction Index (CSI) Method

melzan sabri sabri — 2025-06-08

Canva and slidego are graphic design platforms that are widely used by students. This study uses the Customer Satisfaction Index (CSI) method to measure and understand how satisfied students are with the Canva and Slidego presentation design platforms. This is based on the fact that these platforms are the main choice for students to create attractive professional presentations. This study uses a quantitative, descriptive method by collecting data from students who use both applications through an online questionnaire. To measure overall user satisfaction, but there are also important attributes that affect user satisfaction. such as ease of use, completeness of features. The results of the calculations carried out using the (CSI) method show a value of the percentage of user satisfaction of 85.4115773 Based on the CSI scale, the results are categorized as "very satisfied" with the Canva application. While the Slidego application gets a value of 80.920309% categorized as "Satisfied", so in the results of this CSI calculation, the Canva application has a value greater than the Slidego application value, so the conclusion is that the Canva application is more and more often used by users.

Classification of Matoa Fruit Ripeness Levels Using PCA and KNN Methods Based on RGB Color

Rendika Efando — 2025-06-08

Matoa fruit or in scientific language Pometia Pinnata is a fruit that is quite popular and is spread in several regions. Matoa fruit has several benefits such as for health, processed drinks, and for consumption. However, up to now, matoa fruit farmers are still sorting the quality of ripe matoa fruit using methods that are still manual, which can cause errors and mistakes in sorting the quality of matoa fruit. Based on this problem, this research developed a system that is capable of classifying the maturity level of matoa fruit using RGB feature extraction and using the Principal Component Analysis (PCA) and K-Nearest Neighbor (KNN) methods. This research uses a dataset of 67 data, namely 42 training data and 25 test data consisting of 3 classes, namely raw, mature, and fully cooked. Data classification that applies KNN with the nearest neighbor value, namely K=3, gets an accuracy result of 92%, with 23 image being classified correctly and 2 image being classified incorrectly.

E-Commerce Web Design at XYZ Store with Recommendation System Features

Hafid Roihan — 2025-06-08

This research aims to design and implement a web-based e-commerce system for XYZ Store by providing a fresh vegetable product recommendation feature, using the Apriori algorithm and the Rapid Application Development (RAD) approach. The design process begins with collecting data through interviews to find out the needs of the system. Transaction data for one month was analyzed using the Apriori algorithm to identify patterns of association between products that are often purchased together. The results of this analysis were used as the basis for creating recommendation features. The RAD approach was used because it allows the system to be developed quickly and gradually by involving users in each phase of development. This research analyzes the usability level of the platform using the System Usability Scale (SUS) method. Testing was conducted on 15 participants who had tried the platform prototype in the designed usage scenario. The measurement results indicate that the average SUS score is 78.2, placing it in the “Good” category. The recommendation system feature received a positive response, with the majority of users finding it helpful in finding relevant products. These findings suggest that integrating the recommendation system not only enriches the platform's features but also contributes to improving the overall user experience.

Evaluation of DAPODIK Information Technology Usage Using the DeLone and McLean Model

Achmad Rofiqi Arwin — 2025-06-13

This study aims to evaluate the Basic Education Data System (DAPODIK) based on the DeLone & McLean model using a descriptive statistical analysis approach. The evaluation results show that the Information Quality dimension received the highest score of 86.44%, reflecting users' positive perceptions of the system's data accuracy, completeness, and relevance. The Service Quality dimension scored 84.40%, indicating good responsiveness and technical support, although there is still room for improvement in communication and service continuity. The Use dimension reached 85.33%, suggesting that the system is extensively utilized in educational operational activities. User Satisfaction obtained a score of 82.83%, the lowest among the five dimensions, yet it still reflects a high level of satisfaction, with potential for improvement in technical and user experience aspects. Meanwhile, the Net Benefits dimension scored 85.33%, signifying that the DAPODIK system provides positive impacts on work efficiency, data accuracy, and administrative coordination. These findings indicate that overall, the DAPODIK system performs well and offers tangible benefits, but continuous improvements are needed to optimize user experience and system effectiveness.

Design and Build of an NFT-based Digital Certificate Management Application on the Solana Network

Matthew Imanuel Sumita — 2025-06-13

This research aims to design and build a Non-Fungible Token (NFT) based digital certificate management application on the Solana blockchain network. This application enables users to create, verify, and transfer certificates in the form of secure and decentralized non-fungible tokens (NFTs). Testing was conducted in two main aspects: functional and performance. The functional testing results showed a 100% success rate, indicating that all the application's main features performed as expected. In terms of performance, NFT creation took an average of 29 seconds for 0.00008 SOL, NFT verification took 1.5 seconds at no cost, and NFT transfer took 19 seconds at the same cost as creation. These results show that the application is able to perform its functions efficiently and economically on top of the Solana network. This research contributes to the utilization of blockchain technology for transparent and secure digital document management.

Analysis of IT Governance Implementation using COBIT 5.0 and COBIT 2019

Denny Ganjar Purnama — 2025-06-16

The rapid development of information technology (IT) has affected many sectors, including higher education. Universitas Pembangunan Jaya (UPJ) implemented two versions of COBIT, COBIT 5.0 and COBIT 2019, to improve their IT management. This research aims to compare the effectiveness of the two versions in supporting IT management at UPJ. The research method used is a qualitative approach with a case study involving in-depth interviews, direct observation, and documentation analysis. The data obtained was analyzed using thematic and comparative approaches. The results showed that COBIT 2019 is more flexible, allows customization to the specific conditions of the university, and improves collaboration between departments. Meanwhile, COBIT 5.0 tends to be more rigid and less adaptive to rapid changes. This research concludes that COBIT 2019 is more effective in IT management at UPJ, as it provides greater flexibility in the management of IT resources and response to environmental changes.

Development of a Web-Based Point of Sale (POS) Information System Using the Agile Method

Agnes Monika Halawa — 2025-06-16

Perkembangan teknologi informasi mendorong pelaku usaha untuk mengadopsi sistem digital dalam kegiatan operasional, salah satunya dalam proses penjualan. Penelitian ini membahas perancangan dan pengembangan Sistem Informasi Penjualan (POS) berbasis website menggunakan metode Agile serta pengujian performa sistem melalui performance testing. Metode Agile dipilih karena sifatnya yang iteratif dan fleksibel, memungkinkan pengembangan sistem secara bertahap dan adaptif terhadap perubahan kebutuhan pengguna. Pengujian performa dilakukan dengan menggunakan Apache JMeter untuk mensimulasikan berbagai skenario pengguna, seperti transaksi serentak, login massal, dan beban sistem jangka panjang. Hasil pengujian menunjukkan bahwa sistem mampu menangani hingga 200 pengguna aktif secara bersamaan dengan waktu respon rata-rata di bawah 3 detik dan tingkat error yang rendah. Namun, pada skenario spike 300 pengguna, terjadi peningkatan waktu respon dan sedikit error, menunjukkan perlunya optimalisasi lebih lanjut. Secara keseluruhan, sistem dinilai layak digunakan dan cukup stabil dalam lingkungan operasional skala kecil hingga menengah.

Implementation of Myers-Briggs Type Indicator Approach in Android-Based Counseling Guidance Information System for High School

Andi Mulawati Mas Pratama — 2025-06-16

Guidance and counseling services at SMA Negeri 3 Gorontalo Utara play a crucial role in helping students recognize their potential, interests, and the problems they face. This study aims to develop an Android-based guidance and counseling information system by integrating the Myers-Briggs Type Indicator (MBTI) approach as a tool for identifying students' personality types. The system is designed to present the MBTI questionnaire, automatically analyze the results, and provide counseling recommendations that correspond to each student's personality type. Testing was conducted on 60 eleventh-grade students at SMA Negeri 3 Gorontalo Utara. The results show that the system is capable of effectively classifying personality types and delivering relevant recommendations. The most commonly found personality types were INFP, ESTJ, and ISFJ. Both students and guidance counselors responded positively to the system's usability, ease of use, and clarity. Although limitations remain in terms of advanced personalization and data integration, the approach has proven to enhance the quality of counseling services in a more adaptive and technology-based manner. This study recommends further feature development and system integration into the school's academic management

Designing a Website-Based Dentist Registration Service Application Using Waterfall and Regression Methods

Siti widiyati malihah — 2025-06-16

The development of information technology has encouraged digital transformation in the health care sector, one of which is through a website-based dentist registration service application. J&J Dental Clinic is a clinic located in Jakarta and was established to support the health aspects and dental health services for the general public. This research aims to design and test the system using the Waterfall development method and Regression Testing. The Waterfall method was chosen because it is able to provide a structured and systematic development flow, while Regression Testing is used to ensure the stability of the system after adding or modifying features. Testing was conducted by 200 users on five main features of the system, including new patient registration, login/logout, reservations, financial reports, and admin data management. The test results showed that the average system success rate was above 97%, with the admin data management feature achieving 100% success and the financial report feature having the lowest success rate of 95%. These results prove that the application functions well and is stable, but still has room for improvement on some features. These findings can be the basis for further development of a reliable and user-friendly digital-based healthcare system.

Analysis and Implementation of Mikrotik-Based Access Points Using Hubs with the PPDIOO Method

rahmat rahmat azis — 2025-06-18

This research discusses the analysis and implementation of Mikrotik-based access points using hubs as a connection medium, with the PPDIOO (Prepare, Plan, Design, Implement, Operate, Optimize) methodology approach. PPDIOO was chosen because it is able to provide a systematic, efficient work structure, and can be evaluated in stages using success indicators (KPIs). Tests were conducted at the Operate and Optimize stages with a focus on throughput (Tx/Rx) metrics, average data transmission speed, and packet loss. The test results show a current Tx throughput value of 13.2 Gbps, a 10-second average of 12.9 Gbps, and a total average of 13.7 Gbps, with no packet loss (0 packet loss). Converting these results into percentages shows success above 90% across all indicators, with connection reliability reaching 100%. These findings prove that the PPDIOO method is effective in designing and managing a stable, efficient, and high-performance access point network.

Optimizing Digital Promotion Decision Making with Data-Driven Marketing

Vina Avianingsih — 2025-06-18

In today’s digital marketing landscape, organizations must optimize promotional strategies using data-driven insights. This study proposes an integrated framework combining machine learning (ML) and multi-criteria decision-making (MCDM. Using 1,200 campaigns across platforms, we applied and evaluated regression models—Linear Regression, Random Forest, and Gradient Boosting—to predict ROI from key attributes. Gradient Boosting performed best (R² = 0.82), identifying engagement score, conversion rate, and click-through rate (CTR) as top factors. Used the Analytic Hierarchy Process (AHP) to prioritize campaigns based on predicted ROI, engagement, CTR, and cost-per-click (CPC). This combined approach supports marketers in balancing data-driven accuracy with managerial judgment. A/B testing showed the model-informed group achieved a 17.6% higher ROI and lower CPCs than baseline strategies. This research advances marketing analytics by merging advanced ML with structured decision-making, providing a replicable method for enhancing promotional effectiveness. The findings highlight the significance of behavioral metrics in predicting success and the value of integrating algorithmic precision with human evaluation. This blended approach empowers marketers to move beyond single-metric optimization, enabling more informed and impactful digital marketing strategies in competitive environments.

Comparative Analysis of Desktop and Web-Based Workshop Systems Using the User Acceptance Testing Method

Muhammad Surya Affandi — 2025-06-19

This study aims to compare desktop-based and web-based workshop information systems at PT. Motoreko Mobilindo using the User Acceptance Testing (UAT) method. The evaluation was conducted based on five key indicators: ease of use, access speed, reliability, security, and user satisfaction. Data were collected through questionnaires filled out by active users who had experience with both systems. The results show that the desktop system excels in terms of access speed (96%), reliability (97%), and security (99%). Meanwhile, the web system outperforms in ease of use (97%) and user satisfaction (96%). These findings indicate that although the desktop system has more stable technical performance, the web system is more widely accepted by users due to its convenience and flexibility. This study recommends a hybrid approach as a strategic solution, by maintaining the technical strengths of the desktop system while integrating the accessibility benefits of the web system. Thus, the company can develop a balanced, adaptive information system that aligns with the needs of digital transformation.

Performance Comparison of Random Tree, K-NN, and A-NN Algorithms for DDoS Attack Detection on Software Defined Network (SDN)

Akbar Pandu Segara — 2025-06-19

Software-Defined Networks (SDNs) with a centralized architecture are vulnerable to Distributed Denial of Service (DDoS) attacks, which can cause widespread network service failures. This study aims to compare the performance of three Machine Learning algorithms—K-Nearest Neighbor (K-NN), Artificial Neural Network (ANN), and Random Tree—in detecting DDoS attacks in an SDN environment. The DDoS-SDN dataset, consisting of 104,345 rows and 23 columns, was used with a data split of 70% for training and 30% for testing. Evaluation was conducted using accuracy, precision, recall, F1-score, and AUC-ROC metrics. The results showed that ANN achieved the best performance with an accuracy of 96.85%, precision of 94.35%, recall of 97.79%, F1-score of 96.04%, and AUC of 0.994, followed by K-NN with an accuracy of 88.89% and Random Tree with the lowest accuracy of 86.49%. The superiority of ANN is attributed to its ability to capture complex non-linear patterns, perform automatic feature extraction, and adapt to the heterogeneity of data from the 22 features used. These findings indicate that ANN is the optimal choice for implementing a real-time DDoS attack detection system in an SDN environment, providing a strong foundation for the development of intelligent and adaptive Machine Learning-based network security systems

Integration of Multi-criteria Decision Making Methods for Tobacco Land Assessment: A Case Study Using SMART, TOPSIS, and AHP

Dony Bahtera Firmawan — 2025-06-21

The results of decision making play an important role in achieving a goal in solving certain problems. Decision making process requires data or supporting evidence that can be used as a guide for the selection of solutions based on available alternatives, to produce choices that can increase productivity. Multi Criteria Decision Making (MCDM) method for the analysis of research data namely SMART, TOPSIS, and AHP. The three methods are tested, because each MCDM method has a different way of working or algorithm, so it is necessary to experiment with certain cases. This study aims to determine the performance of the SMART, TOPSIS, and AHP methods with a case study of selecting of tobacco land recommendations. The application of three MCDM methods for alternative analysts of prospective tobacco land based on testing to determine the accuracy of comparing the results/output of the system with expert recommendation solutions using a sample of 10 tobacco land that produce priority/ranking for tobacco land recommendations, shows that the performance of the three methods produces priority selection results different, with an accuracy of SMART 80

Application of Optical Character Recognition (OCR) in Retrieving Book Title Text for Library Archives Digitization

Julian Mulyadi — 2025-06-21

The advancement of information technology requires libraries to transform into digital services. One important step in this process is the digitization of book titles, which has previously been carried out manually and is prone to errors. This study aims to develop a system for extracting book title text from cover images using the Optical Character Recognition (OCR) method based on MATLAB 2017b. The OCR method used in this research applies a template matching and feature extraction approach, where characters are recognized by matching them against built-in character templates in the system after going through an image preprocessing phase. The preprocessing stages include Region of Interest (ROI) selection, grayscale conversion, contrast enhancement, noise removal, and image resizing. After preprocessing, the text is extracted using OCR and stored in digital format. The system was tested using 60 book cover images from the Regional Library of Bengkulu Province, featuring various font types, colors, and lighting conditions. The test results, evaluated using a confusion matrix, show good system performance with an accuracy of 81.67%, precision of 83.05%, and recall of 98.00%. The high recall value indicates that the system is capable of accurately recognizing most of the book title text. Therefore, this system can serve as an initial solution to support the automatic, fast, and efficient digitization of library archives.

NPK (Nitrogen, Phosporus and Potassium) Level Detection Tool in Palm Leaf Organic Fertilizer based on Internet of Things (IoT) using Soil NPK SensorIntegration of Multi-criteria Decision Making Methods for Tobacco Land Assessment: A Case Study Using SMART, TOPSIS, and AHP

Hesmi_Aria Yanti — 2025-06-23

Significant decline in palm oil production; this is due to the termination of fertilizer subsidies and price increases, especially NPK fertilizers. To test the nitrogen (N), phosphorus (P), and potassium (K) content and water content in organic fertilizers made from oil palm leaves, cow dung, water, and other materials. Laboratory testing using chemical analysis and a composting process for 3 months is required. The results of the analysis of N content were 1.63%, P₂O₅ was 0.16%, K₂O was 0.12%, and water content was 12.71%. These results are in accordance with the quality standards of organic fertilizers. The implementation of the NPK content detection tool in oil palm leaves uses these results as a reference. Then the implementation is carried out directly at the location by taking samples of oil palm leaves, cow dung, and water, and then testing is carried out using an integrated NPK detection tool in real time. Where the test results are N = 97 mg/kg, P = 28 mg/kg, and K = 113 mg/kg, these values have not met the minimum threshold for organic fertilizer quality, so it is necessary to add other materials and incubate composting for 3 months or use EM-4 as a liquid that can help accelerate composting. The tool has an accuracy of ±2% and a resolution of up to 1 mg/kg, so the tool is suitable for use as a tool for detecting NPK levels.

Application of Artificial Intelligence for Course Information Chatbot SMK Informatika Komputer Ampana Kota

Anisa Yulandari — 2025-06-24

Ampana City Computer Informatics Vocational School students have problems with enthusiasm for learning at school due to the lack of motivation provided. As a result, many students are lazy to listen or follow class subjects. From the description of the existing problem, the researcher tried to create information media to provide motivation to students which is related to the benefits and goals of learning, using Artificial Intelligence which is formed in a subject information chatbot. The Naïve Bayes Classifier method is used to build chatbots. Assuming independence between features, Naive Bayes is able to work well on text-based classification such as processing questions and answers in chatbots. So you can answer students' questions accurately. The research results show that the use of chatbots contributes significantly to increasing learning motivation, as shown by the Paired Sample T-test of 0.05, indicating that there is a significant increase in student interest in learning. In addition, system testing shows an accuracy rate of 100% in providing responses to student questions. Artificial intelligence-based chatbots can be an effective innovation in supporting learning and increasing students' interest in learning in technology-based educational institutions.

Wireless Local Area Network Analysis Using Quality Of Service and Reliability Maintainability Availability (RMA) Parameters

Ester Agung Natanael — 2025-06-30

Wireless Local Area Network (WLAN) has become an important infrastructure in supporting connectivity in various sectors, but its quality and reliability often experience problems. This study aims to analyze the performance of WLAN networks using two main approaches, namely Quality of Service (QoS) parameters which include throughput, jitter, and packet loss, and the Reliability, Maintainability, and Availability (RMA) method. Tests were conducted before and after network improvements using analysis tools such as Wireshark and iPerf. The results showed that after the repair, throughput increased significantly from an unstable value to an average of 5-5.5 Mbps, jitter decreased consistently, and packet loss reduced from a maximum of 8% to below 2%. In addition, the network availability value increased along with a decrease in repair time (MTTR) and an increase in time between failures (MTBF). This study concludes that the combination of QoS and RMA evaluation provides a comprehensive overview of network performance and reliability, and is effective in supporting decision making for continuous improvement of network systems.

Application of Zachman Framework in the Design of Drug Management Information System Architecture of Purbalingga Health Office

Annisa Dwi Risqi — 2025-06-30

The application of the right information system architecture framework is very important in ensuring efficiency and consistency in data management, especially in drug management information systems in the government sector. This study aims to evaluate the maturity level of the application of the Zachman Framework in the design of the Drug Management Information System at the Purbalingga Health Office using the Capability Maturity Model (CMM) approach. The evaluation was conducted on six main perspectives of Zachman: Planner, Owner, Designer, Builder, Subcontractor, and Functioning System. The measurement results showed the following maturity percentage values: Planner (60%), Owner (40%), Designer (60%), Builder (40%), Subcontractor (20%), and Functioning System (60%), with an average maturity level of 46.67% or equivalent to Level 2 (Repeatable). These results indicate that the design process already has a basic pattern but has not been thoroughly documented or standardized. The main recommendation in this research is to improve technical documentation, HR training, and integration between architecture perspectives to achieve a higher level of maturity.

Model of Computer Network Modeling Based on LAN and WLAN in Shipping Companies Using Methodology of PPDIOO

Zone Tryando Gemenio Brahmana — 2025-06-30

Network models are the primary infrastructure supporting smooth operations, ranging from managing shipment data to inter-branch communication and real-time package tracking. With an optimally designed network, companies can maintain stable connections between devices and systems, accelerate information processing, and enhance responsiveness to customer needs. This study aims to develop and optimize the LAN (Local Area Network) and WLAN (Wireless Local Area Network) at PT APZ using the PPDIOO (Prepare, Plan, Design, Implement, Operate, Optimize) method adapted from Cisco’s approach. Considering the limitations in adding network devices, the study designs an efficient and reliable network using hardware such as the ASUS RT-AC68U router, Procurve Switch 1800-24g, Engenius EAP300 wireless devices, and Vascolink UTP (Unshielded Twisted Pair) CAT 6 cables. Network simulation and design were conducted using Cisco Packet Tracer version 8.2.2 on a system powered by an Intel Core i3-6006U processor with 8 GB of RAM (Random Access Memory) and Windows 10 Pro 64-bit operating system. Test results showed a significant improvement in network performance, with a download speed of 33.82 Mbps, upload speed of 33.38 Mbps, and a low ping response time of 7 ms, indicating a stable network operation.

Development of Employee Attendance Recapitulation Features in Company e-HRD Applications Using the Waterfall Method

Akbarakah Anzalna — 2025-06-30

This study aimed to design and develop an attendance system for employees at PT. Nusrabil Grafika, including those working in the office (Work From Office/WFO) and those working from home (Work From Home/WFH). The main limitation of conventional attendance systems was their dependence on physical location, which restricted attendance recording to employees present at the office. This made it difficult to manage the attendance of employees working remotely and reduced operational efficiency, especially in companies with complex organizational structures and a large number of employees. The research followed the system development phases, including literature review, data collection, needs analysis, system design, implementation, and integration and testing. The hardware used in the system development was an Asus TUF Gaming laptop, while the software used included Sublime Text for coding and Visual Studio Code for debugging and team collaboration. The system was designed to manage and monitor employee attendance in real-time, providing various features such as employee data management, attendance recording, absence limit warnings, and attendance reports.

Development of Web-Based Cash Management System (CMS) Based on Cash Processing Center (CPC) at PT. XYZ

Andriana Suhendi — 2025-06-30

This research aims to develop a web-based financial management system to optimize cash management at PT. XYZ. The background of this research is based on the importance of utilizing technology to support the operations of cash management service companies, such as PT. XYZ, which handles Cash In Transit (CIT) and Cash Management. With the rapid advancement of technology, an application is needed that can simplify cash management processes, improve efficiency, and reduce the risk of errors in recording daily transactions. The research methodology used is qualitative, involving active participation, observation, interviews, and documentation. The researcher directly participated in the application development process with the CPC division for 3 months, observing workflows and system requirements, and discussing with employees and company management. Interviews were conducted to understand managerial needs, while observations focused on monitoring existing cash management operations as the basis for developing a new system. The research results show that the development of the web-based financial management system successfully optimized the recording of cash inflows and outflows, facilitated customer data management, and ensured operational efficiency.

Website Application Development for Web-Based Warehouse Material Management

Rivky Kurniatama — 2025-06-30

Efficient warehouse management is a critical factor in maintaining smooth business operations. However, many companies still use manual methods that are prone to errors and time-consuming. This research aims to design and implement a web-based warehouse management system application that simplifies stock management, records transactions of goods entering and leaving the warehouse, and provides real-time reporting. The research adapts the waterfall method, which includes requirements analysis, system design, implementation, deployment, and testing. This study analyzes the warehouse management issues at Sari Rasa Group as a case study. The application is developed using PHP programming language and MySQL database, with a responsive interface for easy access across various devices. Based on the analysis, the use case diagram in this system includes one actor, the admin. The admin can perform activities such as logging in, managing product data, managing request data, managing goods expenditure data, managing goods input data, and creating daily and monthly summaries. The warehouse management website application will be developed with six classes: user, product data, stock input, output data, stock output, and reports. The test results show that the application can assist in the process of recording goods in and out, enabling warehouse management to minimize user errors.

Pengembangan Sistem Monitoring dan Pelaporan Distribusi Barang Menggunakan Metodologi Waterfall dan Blackbox (Studi Kasus: PT. Mitra Mega Sinergi)

Jeli Sukaramah Zai — 2025-06-30

The digitalization era demands structured management of goods distribution data to improve performance and company competitiveness. This research aimed to develop a goods distribution data management system at PT. Mitra Mega Sinergi. By adopting the waterfall methodology, which consisted of a literature study followed by analysis and design of a web-based application, application implementation, blackbox testing to ensure functionality, and report writing as documentation of the system development results. The case study conducted at PT. Mitra Mega Sinergi identified issues in data distribution management, including data inaccuracies, tracking difficulties, and slow decision-making. Based on the research findings, the system consisted of a database to record user data, goods, and distribution processes. The goods class recorded the stock of goods, while the distribution class managed shipment transactions, including the user, the shipped goods, the destination, and the transaction time. The final report of this system consisted of reports such as the name of the goods, item code, type, unit, available stock, and stock in the warehouse, which were used to identify and monitor the availability of goods.

Application ConTrack for Inventory Management and Monitoring in a Construction Service Company

Siska Mailana — 2025-06-30

This research aims to develop ConTrack application, a goods and transaction management system for PT Alpha Prime Creation. The research methods used include direct observation and interviews with management and warehouse staff to identify problems in manual recording that result in input errors, recording delays, and report inconsistencies. The ConTrack application was designed using waterfall modeling. Functional testing results show that most of the main features, such as input and update of stock data, monitoring of incoming/outgoing goods, and searching for goods, work well with an average success rate of 95%. Performance testing results show that the application has an average response time of 1.2 seconds, can handle 100 active users simultaneously, and process up to 10,000 entries in less than 5 seconds. With these results, the ConTrack application is declared feasible to be applied in the operations of construction service companies as an efficient and reliable stock management support system.

Application NotaryReq for Notary Service Data Management Using Prototyping Methodology

Muhtar Muhtar — 2025-06-30

This research aimed to develop a web-based data management application for notary/PPAT administration at the Notary and PPAT Office of H. Arief Afdal, SH, M.Kn., located in South Jakarta. The software engineering approach used in this study was the prototyping method, where system development was conducted iteratively based on direct user feedback. The research took place from April to June 2024. The system design utilized Unified Modeling Language (UML) with the help of Modelio version 4.1 software. The application was developed using PHP for the backend, JavaScript for the frontend, along with Visual Studio Code and Laragon as the development environment. The developed notary/PPAT data management application included various features with actors such as clients, notary staff, financial staff, and administrators. Clients could submit, verify, and manage request data and payments, while notary staff were responsible for verifying requests, managing various administrative books such as the ledger, land books, and legalization books, as well as monitoring audit logs. Financial staff verified payment data, and administrators had full access to manage all request data, payments, and other administrative books, ensuring data accuracy and monitoring transparency through audit log features.

Integration of Feature Modules for Electric Power Management in Financial Information System (SIK)

Nur Fahrurozi — 2025-06-30

This study aimed to develop a web-based application module to monitor electricity consumption on network devices at PT. LinkNet, an internet service provider. The main issue faced was internet service disruptions caused by the depletion of electricity credit on prepaid meters, which could result in network devices losing power. To address this, a prepaid electricity monitoring module was developed to estimate the time until the credit runs out and prevent service disruptions. The methodology used in this research was the waterfall software development model, which included stages such as literature review, problem identification, data collection, application design and testing, and reporting the test results. The results indicated that the application helped users monitor electricity consumption in real-time, display the estimated time until the electricity credit expired, and provide downloadable usage reports. The application could be used for electricity management and minimizing the potential for internet service disruptions caused by electricity usage management issues.

Dashboard Application for Monitoring Parking Locations and Users Based on the MVC Concept

Abd Rahman — 2025-06-30

The research aimed to develop a web-based parking dashboard application capable of integrating transaction data, membership management, parking slot monitoring, and real-time income reporting. The study was conducted at two parking management locations of PT. Oxient Konsultan Indonesia, namely True-Blue Hotel Cikini (Jakarta) and Intermark BSD (Tangerang). The research lasted for six months, from September to February 2025, with stages including literature study, application requirements analysis, application design, code development, testing, and reporting. The application was developed using CodeIgniter 4 for application structure management, Bootstrap 5 for the user interface design, MySQL with phpMyAdmin for database management, and Chart.js for data visualization. Development was carried out in Visual Studio Code with XAMPP as the local server, using hardware with an Intel Core i5-1135G7 processor and 16 GB RAM. Based on the research findings, the parking management dashboard application featured login options for admins and operators, as well as membership and transaction management. Admins could add new members, while operators were only able to input vehicle entry and exit transactions. The application provided daily reports and a statistical dashboard to display vehicle and income summaries.

WEB PLATFORM DEVELOPMENT FOR OUTSOURCING SERVICES AND EMPLOYEE MANAGEMENT

Chinoble Cofely — 2025-06-30

PT. Buntariva Mitra Solusi, which is engaged in providing services, faces significant problems due to the use of manual systems in managing service requests, employee data, and evaluating service provider performance, which causes delays in the process and decreases the level of client trust. This study aims to design and implement an integrated web platform that is able to automate and unify all of the company's main business processes, including service request management, employee data management, contract arrangements, and real-time monitoring and evaluation of service provider performance. The development method applied is the Software Development Life Cycle (SDLC) with a waterfall model, which allows for systematic and structured project management. The platform implementation is carried out using the PHP programming language, MySQL database, and Apache as a web server. System testing is carried out using the Black Box method to ensure that all functions run according to specifications without any significant errors. The test results show that the system can increase transparency, speed up operational processes, and provide easy access to information for management and clients.

Sentiment Analysis of Lecturers' Pedagogical Competence Using Word Embedding and Random Forest on Student Feedback Data

Vina Ayumi — 2025-06-30

Perkembangan teknologi dan media digital telah mendorong cara evaluasi kinerja dosen yang tidak hanya berbasis kuantitatif, tetapi juga didukung oleh analisis data kualitatif. Salah satu pendekatan yang efektif adalah analisis sentimen terhadap umpan balik mahasiswa terkait informasi berharga mengenai persepsi dan pengalaman terkait kompetensi pedagogi dosen. Penelitian ini mengolah dataset sentimen umpan balik mahasiswa di Universitas Sjakhyakirti dengan menggunakan metode word embedding (WE) dan random forest (RF) untuk mengklasifikasikan sentimen positif dan negatif. Dataset yang digunakan berjumlah 6.164 data kuesioner, terdiri dari 3.800 data sentimen positif dan 2.364 data sentimen negatif. Data kemudian dibagi menjadi data pelatihan (70%), validasi (10%), dan pengujian (20%). Evaluasi kinerja model pada data pengujian menunjukkan bahwa model WE-RF mampu mengklasifikasikan sentimen dengan tingkat presisi, recall, dan F1-score masing-masing sebesar 0,805; 0,724; dan 0,762 untuk kelas positif, serta 0,618; 0,719; dan 0,664 untuk kelas negatif. Akurasi pada tahap pengujian yang diperoleh mencapai 72,2% yang menujukkan bahwa model ini cukup efektif untuk menganalisis sentimen dalam konteks kompetensi pedagogi dosen.

Word2Vec and LSTM-Based Deep Learning Model for Lecturer Professional Competence Feedback Classification

Vina Ayumi — 2025-06-30

This study aimed to develop a deep learning model based on Word2Vec and Long Short-Term Memory (LSTM) to classify sentiment in student feedback on lecturers' professional competence. Manual analysis of large volumes of evaluation text data required significant time and resources, thus an automated method was needed to assist the sentiment classification process. Word2Vec was used to represent words as fixed-dimension numerical vectors, which then served as input to the LSTM model. The LSTM model was selected for its ability to process sequential data and retain relevant long-term contextual information in the text. The dataset consisted of 6,124 evaluation texts, divided into 3,800 positive and 2,324 negative samples. The dataset was split into training (70%), validation (10%), and testing (20%) subsets. The model was trained for 50 epochs, achieving a training accuracy of 81.20% and a validation accuracy of 77.10%. Evaluation using a confusion matrix on the testing data showed that the model correctly classified 587 positive and 359 negative samples, while producing 106 false positives and 173 false negatives. These results indicated that the combination of Word2Vec and LSTM was effective in classifying sentiment in lecturer competence evaluation texts, with a testing accuracy of 77.2%.

Classification of Social Competence Feedback Text of Lecturers in Higher Education Using Word2Vec and CNN-1D

Vina Ayumi — 2025-06-30

The advancement of artificial intelligence technology supported the development of automatic sentiment classification. This study aimed to develop a deep learning model based on Word2Vec and one-dimensional Convolutional Neural Networks (CNN-1D) to classify the sentiment of textual feedback regarding lecturers’ social competence in higher education. The dataset consisted of 6,124 feedback texts collected from student questionnaires at Universitas Sjakhyakirti. The data were proportionally divided into 70% for training, 10% for validation, and 20% for testing. The developed Word2Vec-CNN-1D model demonstrated performance with a training accuracy of 85.10% and a validation accuracy of 79.10%. During the testing phase, the model achieved an accuracy of 76.2% in classifying the feedback texts into positive and negative classes. Evaluation metric analysis showed that for the positive class, the model attained a precision of 0.827, recall of 0.760, and F1-score of 0.792, while for the negative class, it obtained a precision of 0.679, recall of 0.761, and F1-score of 0.717. The results indicated that the Word2Vec and CNN-1D model was more effective at identifying positive sentiments, whereas the performance for the negative class could still be improved in the classification of textual feedback on lecturers’ social competence.

Classification of Personality Competency Feedback Text in Higher Education Using TF-IDF Feature Extraction and Logistic Regression Algorithm

Vina Ayumi — 2025-06-30

This study aimed to develop and evaluate a text classification model to identify sentiment in feedback on lecturers’ personality competencies at a university using TF-IDF feature extraction and Logistic Regression (LR) algorithms. The data originated from student evaluations of lecturers’ personality competencies at Universitas Sjakhyakirti, consisting of a total of 6,112 texts labeled as positive sentiment (3,700) and negative sentiment (2,412). The dataset was then divided into three parts: training (70%), validation (10%), and testing (20%). The research stages included text preprocessing, which involved data cleaning, letter normalization, and the removal of common words, followed by term weighting using the TF-IDF method and classification using the LR model to categorize texts as positive or negative sentiment. The model was evaluated using accuracy, precision, recall metrics, and a confusion matrix. Experimental results showed that at the 50th epoch, the model achieved a training accuracy of 81.90% and a validation accuracy of 78.30%, while on the testing data, the TF-IDF-LR model reached an accuracy of 75.1%.

Preliminary Study: Evaluation of Communication Service Quality in Academic Environment Using Sentiment Analysis

Desi Ramayanti — 2025-06-30

Google Chat was used in the Faculty of Engineering and Informatics to support academic and non-academic collaboration, with advantages such as real-time, structured communication that is easily accessible by lecturers, students, and administrative staff. This study aimed to analyze the sentiment of Google Chat conversations using the support vector machine (SVM) model to categorize sentiment into positive, negative, and neutral. The data preprocessing process included tokenization, stemming, stop words removal, and text cleaning, using Python with NLTK, Scikit-learn, and Pandas libraries. The study adopted a quantitative approach to analyze sentiment patterns from Google Chat conversations about the quality of services at the Faculty of Engineering and Informatics, as well as experiments to develop and test machine learning models. The preliminary study showed a dataset of 3,735 cleaned data, with invalid or empty elements removed. The data was then categorized into three sentiments: positive, negative, and neutral, based on sentiment analysis of communication text related to services in the academic environment. The next research will focus on selecting the appropriate classification model to provide accurate assessments of the faculty’s service quality.

Penerapan Metode Random Contrast dan Brightness Range Pada Dataset Citra Daun Fitomedisin

Mariana Purba — 2025-06-30

This study aimed to enhance the performance of deep learning models in detecting and classifying medicinal plant leaf images by applying two data augmentation techniques, namely Random Contrast Augmentation (RCA) and Brightness Range Augmentation (BRA). The RCA technique randomly adjusted the contrast of images by calculating the pixel average and modifying each pixel value based on a contrast factor, thereby increasing the variation in image lighting. Meanwhile, BRA randomly altered the brightness of the images to simulate varying lighting conditions. The research process began with the collection of medicinal plant leaf image datasets, which were then divided into three parts: training data, validation data, and testing data. The dataset was then pre-processed to prepare the images before applying the augmentation. Augmentation techniques were employed to enrich the dataset by generating modified copies of images using RCA and BRA techniques. The application of both augmentation techniques resulted in a training dataset of 2,400 images, 300 validation images, and 300 testing images.

Application of Augmentation Method on Pharmacognosy Dataset Using Horizontal and Vertical Flip Technique

Mariana Purba — 2025-06-30

This study aimed to apply image augmentation techniques, namely horizontal flip and vertical flip, to a pharmacognosy dataset to increase the diversity of training data in a pharmacognosy image recognition system. By applying these two techniques, this study focused on finalizing a pharmacognosy image dataset that could be used to train machine learning models. The application of these augmentation techniques improved the accuracy and generalization ability of the model in recognizing pharmacognosy images taken from various viewpoints and orientations. This study used two image augmentation techniques, vertical flip augmentation (VFA) and horizontal flip augmentation (HFA), to expand the pharmacognosy image dataset. Each augmentation technique produced four times the number of modified images from the original images with more and more diverse data variations. With the application of the vertical flip augmentation technique, the training dataset consisted of 2,400 images, a validation dataset of 300 images, and a testing dataset of 300 images, for a total of 3,000 data sets. Similarly, the horizontal flip augmentation technique yielded the same amount of data: 2,400 data points for training, 300 data points for validation, and 300 data points for testing. These two techniques increased the total number of training and testing data points to 3,000.

Implementation of Dataset Augmentation on Ethnofimedicinal Images Using Rotation and Channel Shift Techniques

Mariana Purba — 2025-06-30

This study aimed to increase the quantity and variety of ethnopharmacological image datasets using image augmentation techniques, specifically rotation range augmentation (RRA) and channel shift range augmentation (CSA). The dataset augmentation was conducted to enrich the training data for the development of machine learning models used to recognize medicinal plant images. The RRA technique rotated images by random angles, providing variations in object orientation, while CSA altered the color channel values to simulate changes in lighting and the natural colors of plants. The research process included dataset collection, data preprocessing, application of both augmentation techniques, and division of the dataset into training, validation, and testing data. The results showed that the CSA technique produced 2,400 training data, 300 validation data, and 300 testing data, while the RRA technique produced the same amount of data. Therefore, the total data generated from both augmentation techniques amounted to 6,000 images, which could improve the accuracy and performance of deep learning models in recognizing ethnopharmacological images.

Feature Modeling and User Interface Modeling for Ethnopharmacology Data Management System

Mariana Purba — 2025-06-30

This research aimed to develop software for a pharmacognosy data management system by utilizing a feature modeling approach to map and categorize application features based on user requirements. The feature grouping consisted of mandatory and optional features designed to meet the diverse needs of users. The research methods used included literature study, data collection, requirement analysis, feature modeling, user interface modeling, and evaluation. The results of this study led to the development of a pharmacognosy data management system to manage data on medicinal plants, users, and reviews. The main features developed included user management and plant management, both of which were mandatory features for managing users and medicinal plants. Under Plant Management, there were optional features like plant part management and mandatory features like Plant Usage Management to support the management of medicinal plant usage information. Additionally, image data management was added as an alternative feature for managing images of medicinal plants. Thus, the system was expected to serve as a reference for software development that can adapt to user needs and the ongoing technological advancements.