Jurnal Teknologi Informasi Universitas Lambung Mangkurat (JTIULM)
Not a member yet
142 research outputs found
Sort by
Comparison of Naive Bayes and SVM in Public Opinion Sentiment Analysis on Platform X
Get PDFThe growth of social media has made it the primary means for the general public to express their opinions, including on political and legal issues in Indonesia. One topic that has been widely discussed is the abolition of Tom Lembong and the amnesty granted to Hasto Kristiyanto by President Prabowo Subianto, which has garnered mixed public reactions on the X platform. The purpose of this study is to analyze public sentiment regarding current issues and compare the performance of two machine learning algorithms, Naïve Bayes and Support Vector Machine (SVM), to classify public opinion. Data was obtained through a crawling process of 3,003 tweets, followed by a preprocessing stage that included cleaning, case folding, slang normalization, tokenizing, stopword removal, and stemming. Next, a suitability analysis using the TF-IDF method was conducted before the data was processed by the two algorithms. The results showed that, of the 2,998 valid tweets, 78.6% of public opinion was negative and only 21.4% was positive, indicating a predominance of criticism of the issues discussed. When comparing the algorithms, SVM provided more accurate results with an accuracy rate of 78.66%, while Naïve Bayes only achieved 58%. This shows that SVM is more flexible in analyzing text data with a high level of complexity compared to Naïve Bayes
Network Security Analysis with Hybrid Intrusion Detection System, Firewall, and Attacker Log Visualisation
Get PDFThe current digital era brings convenience to people in various industries, including access to information that can be obtained from various sources on the Internet. However, the freedom of the Internet has also led to an increase in cybercrime, which has become a serious problem. According to a monitoring report from the National Cyber and Crypto Agency (BSSN), Indonesia experienced a total of around 2.4 billion cyberattack anomalies between January 2021 and August 2022. With so many cases, an effective system is needed to detect, prevent, and monitor computer networks. This research applies a hybrid Intrusion Detection System (IDS) system that uses OSSEC and Suricata, and uses Elastic Stack for log management for server monitoring. The results show that this hybrid IDS system is able to detect all types of attacks tested, including port scanning, brute force, SQL injection, and denial of service (DoS). In addition, this system can also block attack access by utilising firewall features such as Iptables. The detection results of the hybrid IDS were successfully visualised using Elastic Stack, demonstrating the effectiveness of the system in improving computer network security
Financial Management Website Design Using a Design Thinking Approach (Case Study Noka Dessert)
Get PDFMicro, small and medium enterprises (MSME) such as Noka Dessert often use outdated methods to document financial transactions, which makes tracking finances in real time impossible and is prone to errors. This study intends to explore the creation of a web-based financial management system with the purpose of automating the task of record-keeping and improving real time reporting using the Design Thinking approach. Focusing on users such as owners and employees ensures enhanced accuracy, efficiency, and improved accessibility of the data. The system allows forser authentication as admin and staff, with rights for data entry, validation, and reporting. Functional testing using the System Usability Scale (SUS) has indicated a satisfactory user experience with a mean score of 76. Further research is recommended towards connecting the system with digital payment platforms, overhauling the inventory system, and conducting regular workshops. Noka Dessert is anticipated to improve financial management in this system, leading to sustainable business growth
The Application of Augmented Reality in Furniture Purchasing and Evaluation Based on the System Usability Scale (SUS)
Get PDFThe furniture industry is undergoing a significant digital transformation, reshaping how consumers interact and make purchasing decisions. Augmented Reality (AR) technology enables users to visualise furniture products in their actual physical spaces before making a purchase, offering a more interactive, realistic, and personalised shopping experience. This study aims to evaluate the effectiveness of AR technology in supporting online furniture purchasing by applying the ADDIE development model and assessing usability through the System Usability Scale (SUS) method. A total of 35 respondents, representing millennial and Gen Z users aged between 18 untill 35, participated in testing an AR-based furniture shopping application. The research findings indicate that the application achieved an average SUS score of 81.5, which falls into the "Excellent" usability category, signifying a high level of user satisfaction and acceptance. The results also reveal that AR improves consumer confidence in product selection by allowing realistic visualisation of furniture items in users' own environments. Therefore, this study concludes that integrating AR technology in digital commerce not only enhances user experience but also provides an effective marketing strategy for furniture businesses to strengthen customer engagement, trust, and purchase decisions in the digital era
Optimising OSPF Routing to Improve LAN Network Performance
Get PDFThe performance of a Local Area Network (LAN) greatly depends on the efficiency of the routing protocol applied. This study aims to optimize the configuration of the Open Shortest Path First (OSPF) protocol to improve LAN performance. OSPF is a dynamic routing protocol based on a link-state algorithm that calculates the shortest path using Dijkstra’s algorithm. The research method employed is simulation using Cisco Packet Tracer software. The network topology consists of multiple routers and end-devices, divided into three OSPF areas, each configured with specific IDs, subnets, and IP addresses. The configuration process includes setting IP addresses, assigning OSPF areas, and testing connectivity using ping and traceroute commands. The results demonstrate that OSPF successfully establishes full adjacency between routers, synchronizes the Link-State Advertisement (LSA) database, and ensures optimal routing paths across devices. This implementation proves that OSPF enhances efficiency, convergence speed, and network stability. The study contributes to the development of small to medium-scale LAN networks requiring optimal and reliable data traffic management. It also provides practical insight for network engineers in designing scalable and high-performance routing configurations
Implementation of Midtrans Payment Gateway in the 81 Coffee Sales Application
Get PDF81 Coffee is a coffee shop business established in Banjarmasin City that faces challenges in its conventional sales system and dependency on third-party marketplaces, which affect profit margins and operational efficiency. This study aims to design and implement a mobile-based coffee sales application to improve transaction efficiency, expand customer reach, and provide a better user experience. The research employed a software development method using the Flutter framework for mobile application development, Firebase as the cloud database, and Midtrans as the payment gateway. The system was tested through functional, integration, and field testing using various devices and network conditions to measure response time, transaction success rate, and data consistency. The results show that the application performs effectively, achieving an average UI response time of 1.2 seconds, payment success rate of 98%, and data consistency of 100%. The integration of Midtrans enables a secure and seamless digital payment process. Overall, the developed system improves operational efficiency and provides a reliable digital sales platform for 81 Coffee’s business operations
Performance Evaluation of Random Forest for Hypertension Risk Prediction
Get PDFHypertension is a major global health concern and a leading risk factor for cardiovascular disease, stroke, and kidney failure. Early prediction of hypertension is crucial because the condition is often asymptomatic in its initial stages and late detection increases the likelihood of severe complications. This study aims to develop and evaluate a predictive model for hypertension using the Random Forest algorithm, a robust ensemble learning method well-suited for medical data classification. The dataset used in this research was obtained from Kaggle and contains 1,985 records with 11 attributes representing demographic, lifestyle, and clinical risk factors. Preprocessing was performed to ensure data quality, followed by Random Forest classification with different parameter settings. The model was evaluated using 5-fold and 10-fold cross-validation with various numbers of trees ranging from 50 to 250. Performance metrics included accuracy, precision, recall, F1-score, and AUC. Experimental results demonstrated that the Random Forest algorithm achieved consistently high performance, with accuracy above 93%, precision above 95%, recall above 91%, F1-scores above 93%, and AUC values between 0.986 and 0.991. These findings confirm that Random Forest is highly effective and reliable for predicting hypertension risk. The study highlights the algorithm’s potential as a decision-support tool for early detection, enabling preventive measures and improving public health outcomes
Design of Website-Based E-Commerce Information System Using Extreme Programming Method (Case Study: Belv Boutique)
Get PDFThe purpose of this research is to provide a REST API-based web service for an online fashion product ordering platform, using butik belv as a case study. The purchase process carried out at butik belv At this time, consumers must visit butik belv directly to complete their purchase because it is currently still processed offline, where payments are still made in cash and buyers must visit the store directly to complete the purchase. Payments and transactions are still paid in cash. In addition, sales transactions are still recorded manually which are prone to errors. This research aims to create a web application that utilizes REST API technology. With the help of this application, users can order products online and pay directly through the website without having to visit the store. In addition, the transaction recording system is automated, thus ensuring that all sales data is recorded accurately, this reduces the possibility of recording errors and simplifies the sales reporting process. Extreme programming is the software development methodology used. It includes design, coding, testing, and planning phases. The system also uses Blackbox testing which is expected the test results show that each functionality works as intended, indicating that the use of web services can increase the speed of the application and streamline the purchasing process and assist buyers in the purchasing process. Butik butik belv anticipates increased operational effectiveness and provides this solution
Evaluation of User Experience in the Game "Night's Reach" Using the Game Experience Method Questionnaire (GEQ)
Get PDFStudents from Lambung Mangkurat University developed a game called "Game Night's Reach" aimed at introducing and preserving wetland environments. To assess the game's viability for release, it is crucial to test the UI/UX to gauge its efficiency. This involves understanding how well players interact with the game and their over-all user experience. The testing is conducted through distributing questionnaires to collect gameplay experience data from new users. The method used is the Game Experience Questionnaire (GEQ), which measures the efficiency of user experience by evaluating players' feelings and experiences while playing the game, beyond just usability aspects like efficiency, perspective, and dependency. Based on the questionnaire data, a redesign was performed in the first phase to improve the game, which will be compared in the second phase. The usability evaluation of "Game Night's Reach" using the GEQ method showed the following changes after redesign: Competence, Increased by 0.198 from 0.872 to 1.068.
Immersion, Increased by 0.196 from 0.853 to 1.051. Flow, Increased by 0.29from 0.742 to 1.032. Tension, Increased by 0.1 from 0.51 to 0.61. Challenge, Increased by 0.13 from 0.588 to 0.718.Negative affect, Decreased by 0.103 from 0.545 to 0.442. Positive affect, Increased by 0.106 from 0.894 to 1.008.Empathy, Increased by 0.235 from 0.723 to 0.958. Negative feelings, Decreased by 0.072 from 0.482 to 0.41. Behavioural involvement, Increased by 0.388 from 0.59 to 0.978. Positive experience, Increased by 0.168 from 0.89to 1.058. Negative experience, Decreased by 0.078 from 0.501 to 0.423. Tiredness, Decreased by 0.01 from 0.505 to 0.495. Returning to reality, Increased by 0.334 from 0.646 to 0.986
Real-Time K-Means Clustering with Firebase ML Kit: Segmenting Barber Shop Customers by Booking Behavior and Churn Risk
Get PDFThis study addresses the challenges of high churn rates and the need for real-time customer behavior analysis in barber shops by developing a K-Means model integrated with Firebase ML Kit. The research analyzes 70,000 booking records from an Android application, focusing on features such as booking frequency, average cancellations, and recency day. The model achieves optimal performance with 5 clusters, validated by a Silhouette Score of 0.58 and a Davies-Bouldin Index of 0.541. Key segments like "Inactive Members" and "High Volume Churners" are successfully identified, enabling targeted business strategies such as reactivation campaigns and priority booking offers. The system is implemented in a mobile application, providing real-time customer segmentation and actionable insights. This approach offers a scalable solution to enhance customer retention and operational efficiency in the barber shop industr