Journal Of Information System And Artificial Intelligence

Evaluation of 32 mm Automatic Hair Curlers on Shopee Using AHP Customer Reviews

2026-05-19T19:54:59+07:00

Abstract

This study aimed to evaluate 32 mm automatic hair curler products available on the Shopee e-commerce platform by applying the Analytic Hierarchy Process (AHP) combined with customer review analysis. The data were obtained through direct observation of three products with similar characteristics. The collected data included product ratings, number of reviews, star-based review distribution, sales volume, and price. Sentiment analysis was conducted by categorizing customer reviews into positive, neutral, and negative groups based on their star distribution. The AHP method was used to determine the priority weights of each evaluation criterion. The results showed that sales volume was the most influential criterion in the decision-making process, followed by product ratings and the number of reviews. The final AHP score revealed that the MAIMEITE product achieved the highest value, making it the most recommended option. This study demonstrated that integrating AHP with customer review analysis provided an objective, systematic, and data-driven approach to product evaluation.

Optimization of the K-Means Algorithm Using PCA Dimensionality Reduction for E-Commerce Customer Segmentation

2026-05-09T15:55:51+07:00

The rapid growth of the e-commerce industry in recent years has generated increasingly large and complex volumes of customer data. This data holds strategic potential to be analyzed in order to understand customer behavior patterns and to support data-driven decision-making. This study aims to identify customer segmentation through an unsupervised learning approach using Principal Component Analysis (PCA) and the K-Means algorithm. The dataset used in this research demonstrates good quality with no missing values, making it suitable for further analysis. Initial exploratory findings indicate that Total Spending, Number of Items Purchased, and Average Rating are the most significant variables in representing customer characteristics. The application of PCA successfully reduced data dimensionality while retaining 79.41% of the total variance, thus producing a more concise representation without compromising essential information. The clustering process using K-Means grouped customers into three clearly distinguishable clusters. The first cluster represents customers with high activity levels, the second cluster reflects customers with moderate activity, and the third cluster corresponds to customers with lower engagement intensity. Validation using the Elbow Method and Silhouette Score confirmed that k = 3 is the most optimal number of clusters. Cluster visualizations show strong separation between groups and consistent relationships among variables. This study demonstrates that the combination of PCA and K-Means is effective in producing informative and interpretable customer segmentation. These findings provide a foundation for subsequent analyses and support data-driven decision-making in e-commerce customer management.

Implementation of a Web-Based Expert System for Machine Damage Diagnosis Using Backward Chaining and Certainty Factor

2026-05-09T15:55:28+07:00

PT Hardo Soloplast is a manufacturing company that produces plastic products using various advanced machinery. A recurring issue in the production process is the sudden malfunction of machines, which disrupts operations and increases repair costs. To address this, this study proposes a web-based expert system for machine fault diagnosis using the Backward Chaining reasoning method and Certainty Factor approach. The system is implemented using PHP, HTML, CSS, and JavaScript, and stores knowledge in JSON format. It is accessible via web browser for field technicians. The system covers five main machines: Extruder Starex 1500, Laminating HL-2000, Printing Roto-Gravure, Slitting Rewinder RS-3000, and Blown Film Extrusion Machine. The knowledge base consists of 65 rules and symptoms, collected from interviews and documentation. Backward Chaining was chosen for its efficiency in goal-driven reasoning, while Certainty Factor is applied using the formula CFcombine = CF1 + CF2 × (1 − CF1), with a threshold of 0.75 for reliable results.Testing was conducted by comparing system diagnoses with actual technician assessments, achieving accuracy between 75% and 92.58% . This system contributes to the digitalisation efforts at PT Hardo Soloplast by accelerating diagnosis, improving maintenance response time, and reducing dependence on manual fault identification.

A Comparative Study of Naïve Bayes, SVM, Random Forest, and LSTM Performance in Sentiment Analysis on a Movie Review Dataset

2026-05-09T15:55:02+07:00

Sentiment analysis is an important task in natural language processing, aimed at identifying and classifying opinions or emotions in textual data. This study compares the performance of four classification algorithms—Naïve Bayes, Support Vector Machine (SVM), Random Forest, and Long Short-Term Memory (LSTM)—on 25,000 English-language movie reviews with balanced sentiment labels. Text preprocessing includes cleaning, tokenization, and TF-IDF vectorization for traditional models. For LSTM, both randomly initialized embeddings and pre-trained embeddings are tested. Results, evaluated using accuracy, F1-score, and confusion matrix, show that SVM performs best with 89% accuracy, followed by Naïve Bayes and LSTM at 86%, and Random Forest at 82%. LSTM performs poorly with TF-IDF or self-trained embeddings but improves significantly with pre-trained embeddings. These findings indicate that traditional models, especially SVM, remain highly effective for sentiment analysis on moderately sized datasets, while LSTM requires proper text representation to perform competitively.

UI/UX Design for School Website Using Design Thinking Method at SMP Negeri 2 Palembang

2026-05-09T15:54:35+07:00

The development of information technology encourages educational institutions to provide digital services that can meet the needs of communication, transparency, and information accessibility. School websites function not only as a medium for conveying academic information, but also as a means of interaction between teachers, students, parents, and the general public. This study aims to design the UI/UX design of the SMP Negeri 2 Palembang website using the Design Thinking method which consists of five stages, namely Empathize, Define, Ideate, Prototype, and Test. This approach was chosen because it is oriented towards user needs and is able to produce innovative, applicable solutions. Data was obtained through interviews, questionnaires, and observations to explore user experiences and problems that arise in using the school website. The analysis results showed problems in aspects of navigation, information accessibility, visual appeal, and website responsiveness to various devices. Based on these findings, a website prototype was designed with the main elements of teacher, student, and admin dashboards, student monitoring features, a learning material repository, an activity gallery, and a more structured announcement system. The trial using the System Usability Scale (SUS) obtained an average score of 89.5 which is included in the "Excellent" category with a grade of "A." These findings indicate that the developed UI/UX design can improve user convenience, effectiveness, and satisfaction, while simultaneously strengthening the school's image in the digital age

Web-Based Information System for Laundry Services: An Extreme Programming Approach to Digital Transformation

2026-05-09T15:54:16+07:00

The rapid growth of digital technologies has accelerated the transformation of small and medium-sized enterprises (SMEs), including service-based industries such as laundry businesses. However, many of these businesses still rely on manual processes, which often result in inefficiencies and limited customer engagement. This study aims to develop a web-based information system for laundry services using the Extreme Programming (XP) methodology to support digital transformation. The development process consisted of four stages: planning, design, coding, and testing, carried out iteratively with continuous stakeholder involvement. Key features implemented include customer registration, order management, real-time order tracking, invoice generation, and administrative reporting. Functional testing confirmed that all core features performed as expected. Usability evaluation using the System Usability Scale (SUS) produced an average score of 82.3, indicating Excellent usability. Customers appreciated the mobile-first design and order tracking, employees highlighted the efficiency of order input, and administrators valued reporting functions. The findings demonstrate that XP enables rapid development, enhances system quality, and aligns software outputs with user needs. This study contributes to the literature by showing how agile methods can accelerate digital transformation in SMEs, while also offering practical implications for improving operational efficiency and customer satisfaction in the laundry service sector

Design and Implementation of a Digital Police Assistance Application for Enhancing Public Services

2026-05-09T15:53:59+07:00

Public service delivery in law enforcement is often hindered by limited accessibility, manual procedures, and lack of transparency in complaint handling. This study proposes the Digital Police Assistance Application, a mobile platform designed to integrate incident reporting, real-time tracking, information dissemination, and an administrative dashboard for police officers. The system was developed using the Waterfall model with Flutter for cross-platform mobile development, Node.js with Express for backend services, and MySQL for data storage. Functional validation through Black Box testing confirmed that all modules operated as intended. Usability evaluation involving 30 participants (15 citizens and 15 police officers) employed the System Usability Scale (SUS). The application achieved an average SUS score of 82.5, categorized as “Excellent,” with citizens scoring slightly higher (84.2) than police officers (80.7). Citizens appreciated the simplicity and transparency of complaint tracking, while police officers highlighted dashboard efficiency and suggested integration with internal databases. These results demonstrate that the application is both functional and user-friendly, with strong potential to enhance transparency, efficiency, and trust in police services. Future work will address system interoperability, security improvements, and pilot testing across multiple jurisdictions

Analysis Of The BPBD Website Of South Sumatra Province Using The Pieces Method

2026-05-09T15:53:35+07:00

The rapid development of information technology has made it easier for people to access various information through the internet. Websites are one of the most effective media in disseminating information to the public. The Regional Disaster Management Agency (BPBD) of South Sumatra Province already has a website that provides various information related to disasters, organizational structure, public services, and publication of activities. However, since the beginning of its creation, the website has not been visited many times and has never been analyzed on its performance. This study aims to analyze the performance of the BPBD website of South Sumatra Province using the PIECES (Performance, Information, Economics, Control, Efficiency, and Service) method. The results of the analysis show that from the Performance aspect, the website obtained a grade C score on Google PageSpeed Insight and a grade D on Pingdom; the Information aspect shows a scale value of 3 (good); the Economics aspect is considered efficient due to low maintenance costs; the Control aspect obtained a grade C on the Sucuri Sitecheck and a grade D on the Observatory; the Efficiency aspect shows quite good results; while the Service aspect is considered quite good but still needs improvement in the presentation of information to the public. Overall, the South Sumatra Province BPBD website functions quite well but still needs optimization in its performance and service quality so that it can become a more effective disaster information medium.

Book Recommendation System Using Similarity-Based Collaborative Filtering Approach

2026-05-11T09:44:23+07:00

This study implements and compares the performance of three primary approaches to book recommendation systems: rank-based methods, similarity-based collaborative filtering (both user-user and item-item), and matrix factorization-based collaborative filtering. The dataset comprises 433,671 user ratings from 78,805 users on 185,973 books, enriched with book metadata such as title, author, publication year, and publisher. The recommendation systems were developed using the Surprise library, with hyperparameter optimization performed via grid search cross-validation. Model performance was evaluated using precision@k, recall@k, and F1-score metrics, as well as RMSE for prediction accuracy. Results indicate that the user-user similarity-based collaborative filtering model achieved the best performance in terms of relevance, attaining an F1-score of 0.86. This model effectively identifies users with similar preferences and recommends books based on collective behavior patterns. Meanwhile, the matrix factorization approach yielded the lowest RMSE value of 1.50, highlighting its strength in capturing latent factors that influence user preferences. The item-item similarity model also showed reasonable performance but did not surpass the other approaches, possibly due to homogeneity in item rating patterns across users. Overall, the study confirms that user-user similarity is highly effective for datasets exhibiting consistent user behavior, while matrix factorization excels in minimizing prediction error by leveraging latent feature structures. These findings offer valuable insights for developing adaptive recommendation systems in book-centric literacy platforms and content-driven e-commerce applications.

Mobile Financial Management Application at Yanto Pulsa Using Flutter

2026-05-19T19:53:31+07:00

The utilization of digital technology in financial management has become an important need for micro, small, and medium enterprises (MSMEs) to improve recording accuracy and facilitate financial monitoring. However, many MSMEs still rely on manual financial recording, which may lead to recording errors and difficulties in data management. This study aimed to develop a mobile-based financial recording application as a solution for financial management at Yanto Pulsa. The system was developed using the Waterfall method, which consisted of requirement analysis, system design, application development, and testing stages. The application was built using the Flutter framework with Firebase as the backend and Cloud Firestore as the database. The results showed that the application was able to record income and expense data in real-time, present financial summaries, and support effective data management and backup. Black-box Testing results indicated that all application functions operated according to user requirements.