Smart Techno (Smart Technology, Informatics and Technopreneurship)

Implementation of a Web-Based Boarding House and Rental Search and Booking Information System Using the Waterfall Method

2026-09-30T21:03:39+00:00

The rapid development of information technology has increased the need for a more efficient and structured housing search system for students. This study aims to develop an integrated web-based information system for searching and booking boarding houses and rental properties in accordance with user requirements. The development method employed is the Waterfall model, which consists of the stages of Requirement Analysis, System Design, Implementation, and Testing. Data were collected through observations of Facebook social media, questionnaires distributed to 81 students, and interviews with 10 boarding house owners. The system was developed using PHP with the Laravel framework and a MySQL database. System testing was conducted using Black-box Testing and User Acceptance Testing (UAT) involving 20 respondents, resulting in a user satisfaction rate of 87.4%. The findings indicate that the developed system enhances the efficiency of housing searches and facilitates the online booking process. With a structured approach grounded in empirical needs, the system is considered effective and feasible as a web-based solution for searching and booking boarding houses.

The Influence of Smart Technology Experience on Satisfaction Mediated by Generation Z Loyalty in Bali's Tourism and Hospitality Industry

2026-09-30T21:03:23+00:00

This study examines how Smart Technology Experience shapes Generation Z's satisfaction and loyalty to the tourism and hospitality industry in Bali, given the acceleration of digital transformation of services and the high expectations of young tourists for technology-based experiences. This study uses a quantitative approach with an explanatory design through a survey of Generation Z respondents who have experience using smart technology-based tourism and hospitality services in Bali. Data were analyzed using Partial Least Squares–Structural Equation Modeling to test the direct relationship between variables as well as the mediating role of satisfaction. The results of the analysis show that Smart Technology Experience has a positive effect on satisfaction, and satisfaction has a positive effect on loyalty. Smart Technology Experience also has a direct effect on loyalty, so that Generation Z's loyalty is not only formed through satisfaction evaluations, but also through the value of technology experiences that are directly felt. The mediation test indicates that satisfaction mediates the relationship between Smart Technology Experience and loyalty partially. These findings underscore the importance of managing consistent and relevant technology experiences to strengthen satisfaction while building Generation Z loyalty in the context of Bali destinations.

Design and Development of the Rescentify Website as a Platform to Support Product Marketing

2026-09-30T21:03:17+00:00

The rapid development of digital technology has encouraged businesses to utilize websites as a primary medium for marketing and information dissemination. Rescentify, a business engaged in environmentally friendly products, requires a digital platform capable of introducing its products, building customer trust, and expanding its marketing reach. Therefore, this study focuses on the design and development of the Rescentify website as a digital information and marketing platform using the Prototype method. This method involves iterative stages, including communication, quick planning, quick design, prototype construction, and delivery and feedback, allowing user requirements to be effectively identified and implemented.The developed website incorporates key features such as a homepage, product catalog, detailed product descriptions, a scent recommendation system (Scent Advisor) based on user mood, article pages, and integrated communication through WhatsApp for transaction purposes. The website design emphasizes a simple and user-friendly interface to enhance usability and accessibility for a wide range of users. In addition, Search Engine Optimization (SEO) techniques and social media integration are applied to improve website visibility and expand digital marketing reach. The results of system testing using the Black Box method indicate that all features function as intended and meet the predefined system requirements. The implementation of the Rescentify website demonstrates its effectiveness as a promotional and information platform, facilitating easy access to product information and improving interaction between customers and the business. Therefore, the website is expected to support more optimal, efficient, and sustainable digital marketing activities, particularly for environmentally friendly products

Business Model Design of a Digital-Based Outfit Recommendation Application for Fashion Personalization

2026-09-30T21:03:32+00:00

The rapid development of the digital fashion industry has increased the demand for personalization in outfit and makeup selection, particularly among Generation Z and millennial women who face time constraints and demonstrate growing awareness of sustainability. Common issues include difficulties in matching clothing according to personal characteristics, low self-confidence due to inappropriate outfit choices, and a tendency toward overconsumption influenced by fast fashion practices. This study aims to design a business model and develop a user interface prototype for the ColorMatch application as an AI-based outfit recommendation platform, utilizing the Business Model Canvas (BMC) framework and the Prototyping development method. A descriptive qualitative approach is employed to analyze the nine elements of the BMC, while the prototyping stages include requirement identification, wireframe design, mockup development, and feedback-based evaluation to produce a user-oriented UI/UX design. The results indicate that ColorMatch offers a value proposition in the form of outfit and makeup recommendations based on seasonal color analysis, mood-based styling, and personal wardrobe optimization. The integration of digital business model approaches, AI technology, and UI/UX design through prototyping results in a platform that is economically viable, responsive to user needs, and supportive of sustainable fashion practices.

Implementation of the Naive Bayes Algorithm for Public Sentiment Analysis Toward Power Plant Development

2026-09-30T21:03:04+00:00

This study aims to analyze public sentiment toward power plant development using data collected from the social media platform Twitter. The dataset consisted of 2,493 tweets obtained through a crawling process using keywords related to power plant development, such as “PLTS” and “PLTS Cirata.” The preprocessing stage included cleaning, case folding, stopword removal, tokenization, and stemming using the Sastrawi library to produce more structured textual data. The dataset was then divided into a training set comprising 85% of the data (2,120 tweets) and a testing set comprising 15% (373 tweets). The classification process was performed using the Multinomial Naive Bayes algorithm, as this method is well suited for text data represented by word-frequency features extracted through CountVectorizer. Model evaluation was conducted using a confusion matrix with accuracy, precision, recall, and F1-score as performance metrics. The results showed that the model achieved an accuracy of 75%, indicating that the Multinomial Naive Bayes method is reasonably effective for text-based sentiment classification. Furthermore, the findings revealed that public opinion regarding power plant development is influenced by perceptions of renewable energy benefits, environmental impacts, and government policies.

Evaluating the Performance of the No Language Left Behind Encoder for Sentiment Classification Across Multiple Indonesian Regional Languages

2026-09-30T21:03:10+00:00

This study aims to evaluate the performance of the encoder from the No Language Left Behind (NLLB) model for sentiment classification tasks across several Indonesian regional languages. Originally developed for machine translation, the NLLB model is explored for its ability to generate contextual text representations that are relevant to sentiment classification. The dataset used in this study is NusaX, which comprises 12 languages, including Indonesian, English, and 10 Indonesian regional languages. Two training approaches were employed: fine-tuning, in which all model parameters were updated using sentiment classification data, and partial fine-tuning, in which only the upper layers were updated while the embedding layer was frozen to preserve the original lexical representations. Training was conducted using the AdamW optimization algorithm with CrossEntropyLoss as the loss function and mean pooling as the feature aggregation mechanism. Model performance was evaluated using accuracy and macro F1-score metrics at both the multilingual and per-language levels. The results indicate that the two approaches achieved comparable performance, with an accuracy of 81% and a macro F1-score of 80% on the multilingual dataset. Per-language analysis further revealed that the model performed better on languages that had been included in the original NLLB pretraining data, such as Acehnese, Balinese, Banjarese, Minangkabau, Javanese, and Sundanese, achieving accuracy scores ranging from 79% to 86%. In contrast, several languages that were not covered during NLLB pretraining, including Ngaju, Madurese, and Toba Batak, exhibited slightly lower performance, with accuracy scores ranging from 70% to 78%. These findings demonstrate that the NLLB encoder possesses strong adaptation capabilities for text classification tasks, even in the context of low-resource regional languages.

UI/UX Design Using the Design Thinking Approach for a Location-Based Attendance System at PT Agung Sumatera Development

2026-09-30T21:02:58+00:00

Digital transformation has accelerated the adoption of technology-based attendance systems; however, manual attendance methods continue to exhibit limitations, including data manipulation and operational inefficiency. This study aims to design an optimal User Interface (UI) and User Experience (UX) for a location-based employee attendance system at PT Agung Sumatera Development using the Design Thinking approach up to the prototyping stage. The design process consisted of the empathize, define, ideate, and prototype phases, incorporating the MoSCoW prioritization framework Must-have, Should-have, Could-have, and Won't-have as well as the How Might We (HMW) technique to identify user needs and generate design solutions. Figma was employed as the primary design tool for developing the prototype. The proposed system integrates GPS-based location verification, real-time selfie authentication, and automatic watermark generation to enhance attendance accuracy and reliability. The resulting prototype features a simple workflow and intuitive navigation. A usability walkthrough involving five employees indicated that the design effectively met usability criteria in terms of ease of understanding and navigation clarity, achieving an average Task Completion Rate of 90%. Feedback regarding the readability of the location radius instruction was incorporated into the final design, resulting in improved user experience.

UI/UX Design Of A Peer Counseling Website For Universitas Mulawarman Students Using The Design Thinking Method

2026-09-30T21:02:52+00:00

This study aimed to design a UI/UX-based website for peer counseling services at Universitas Mulawarman to improve accessibility and user experience. The existing counseling service, known as CIRCLE, had limited awareness and relied on non-integrated platforms, which reduced its effectiveness in meeting students’ needs. A user-centered approach was applied using the Design Thinking method, consisting of empathize, define, ideate, prototype, and test stages. Data were collected through surveys and in-depth interviews to identify user problems and expectations. The design process was carried out using Figma, including the development of wireframes, style guides, mockups, and interactive prototypes. Visual enhancements such as 3D icons were also applied to create a more engaging interface. The prototype was evaluated through usability testing using Maze and a System Usability Scale questionnaire. The results showed that the designed interface improved user understanding, navigation, and overall experience. The evaluation produced an average usability score of 77, indicating that the system was acceptable and easy to use. These findings demonstrated that a user-centered UI/UX design could enhance the effectiveness, accessibility, and usability of peer counseling services for university students.

Business Model Canvas Analysis and Financial Feasibility Assessment of the “TEMATIK” Math Board Game as an Educational-Based Technopreneurship Innovation

2026-09-30T21:02:45+00:00

This study aims to analyze the digital business model, the effectiveness of digital marketing strategies, and the business feasibility of the “TEMATIK” Math Board Game. This study employed a quantitative descriptive approach using a case study method. Primary data were obtained from sales records during seven weeks of operations, capital and production cost data, and expert validation results, while secondary data were sourced from the program’s final report and digital marketing documentation. The analysis was conducted using the Business Model Canvas framework, which comprises nine key business elements, along with five financial feasibility indicators. The results indicate that the TEMATIK product offers differentiation through the integration of a physical board game with an Android-based educational application covering mathematics topics for elementary school. A hybrid marketing strategy utilizing Shopee, Instagram, WhatsApp, and direct visits to elementary schools demonstrated a shift in channel distribution from an initial dominance of offline channels toward increased adoption of online channels, with total sales reaching 70 units during the first seven weeks. Based on the financial feasibility projections, the business is estimated to reach its break-even point at sales of 103 units. The business feasibility analysis yielded a Benefit-Cost Ratio of 1.46, a Return on Investment of 45.7%, a Cost of Goods Manufactured of IDR 58,333 per unit, and a Payback Period of 6.6 months. Overall, the findings confirm that the digital business model of the “TEMATIK” product is financially feasible and has the potential to be further developed as a scalable education-based technopreneurship product.

Design and Implementation of a Web-Based Mobile Phone Sales Forecasting Application Using the Single Exponential Smoothing Method

2026-09-30T21:02:39+00:00

This study was conducted to address the inventory management issues at iCare Apple Store, where mobile phone stock planning has traditionally relied on estimation, leading to the risk of both stock shortages and overstocking. The objective of this study was to design and implement a web-based mobile phone sales forecasting application using the Single Exponential Smoothing (SES) method. The dataset consisted of monthly sales data for the iPhone XR and iPhone XS from January 2021 to December 2024. The forecasting process was performed by testing alpha values ranging from 0.1 to 0.9, while forecasting accuracy was evaluated using the Mean Absolute Percentage Error (MAPE). The results indicated that the optimal alpha value for the iPhone XR was 0.1, yielding a MAPE of 27.04% and a forecast of 11.45 units, which was rounded to 11 units. For the iPhone XS, the forecasting model produced a MAPE of 52.33% with a forecast of 2.45 units, rounded to 2 units. These findings demonstrate that the Single Exponential Smoothing method can be effectively implemented in a web-based application and can support decision-making in determining appropriate inventory levels for subsequent periods.

Evaluation and Redesign of the One Stop Service Telkom University Website Interface Based on User-Centered Design Principles

2026-09-30T21:04:17+00:00

Website One Stop Service (TOSS) of Telkom University is a web-based academic administrative service platform used by students to submit various administrative requests. However, preliminary observations revealed several usability issues related to the user interface. These issues include the use of an icon-based sidebar navigation that is not sufficiently informative, a dashboard with limited functionality, and a submission history feature that is separated by document type, making the navigation process less efficient. This study aims to evaluate and redesign the TOSS website interface to improve user efficiency and experience. The design process employed the User-Centered Design (UCD) method based on the ISO 9241-210 standard, with a focus on the needs of students as the primary users of the system. The evaluation process involved 15 active student respondents through observation, questionnaires, and usability testing. The redesigned interface features a simplified navigation structure, the presentation of primary services on the dashboard, and the consolidation of submission histories into a single page. Based on usability testing of the redesigned prototype, the system achieved a maximum effectiveness level, with a 100% Success Rate across all task scenarios. User satisfaction was further evaluated using the System Usability Scale (SUS). In conclusion, the UCD approach successfully produced a more user-friendly and intuitive interface that better supports the delivery of academic administrative services.

Sentiment Analysis of Public Opinions on Electric Vehicles on Social Media Using the Support Vector Machine (SVM) Algorithm with Optimized Preprocessing

2026-09-30T21:04:11+00:00

The development of regulations and trends in electric vehicle adoption in Indonesia has generated diverse public responses and opinions on social media platforms such as YouTube. However, manually evaluating large volumes of unstructured review data tends to be inefficient and less effective. This study applies the Support Vector Machine (SVM) algorithm with a Radial Basis Function (RBF) kernel, utilizing Term Frequency-Inverse Document Frequency (TF-IDF) weighting and optimized preprocessing techniques to automatically classify public opinion sentiment. The dataset consists of 1,517 Indonesian-language YouTube comments that underwent a series of preprocessing stages, including case folding, cleaning, tokenization, stopword removal, normalization, and stemming. The resulting cleaned dataset was then transformed into a numerical matrix containing 1,180 word features and divided into training and testing sets using a 70:30 ratio. The initial labeling results showed that public opinions were dominated by negative sentiment, accounting for 57.3% (869 comments), followed by positive sentiment at 33.2% (504 comments) and neutral sentiment at 9.5% (144 comments). The performance evaluation of the SVM classification model using a confusion matrix achieved an overall accuracy of 78.1%. These findings demonstrate that the application of optimized preprocessing techniques combined with a nonlinear SVM kernel provides reliable performance in accurately identifying and mapping the characteristics of digital public opinion.

Comparative Analysis of Naïve Bayes and SMOTE-Based Long Short-Term Memory (LSTM) for Electric Vehicle Sentiment Analysis on YouTube

2026-09-30T21:04:05+00:00

The transition to electric vehicles in Indonesia has generated diverse public opinions on social media. Most previous sentiment analysis studies have tended to employ a single classification method without in-depth comparison and have overlooked the issue of extreme data imbalance, which can introduce bias into classification models. Furthermore, a research gap remains regarding the effectiveness of complex deep learning models compared with simpler statistical models when applied to medium-dimensional Indonesian-language opinion texts. This study aims to address this gap by conducting a comparative evaluation of model performance. The dataset consists of 1,512 unstructured public opinion comments collected from YouTube, distributed across three sentiment classes: 868 negative, 503 positive, and 141 neutral comments. To address the class imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) was applied. The study then compares the statistical Naïve Bayes method using TF-IDF weighting with the Deep Learning Long Short-Term Memory (LSTM) method using Word Embedding. Evaluation using a Confusion Matrix demonstrates that Naïve Bayes outperformed LSTM and exhibited greater stability in sentiment classification for this dataset, achieving an accuracy of 63.04%, compared with 57.43% for the LSTM model.

Sentiment Analysis Using Machine Learning for Evaluating Public Perception to Support Accountability in Vocational Higher Education

2026-09-30T21:03:59+00:00

This study examines public sentiment toward vocational higher education in Indonesia using 2,014 YouTube comments from 79 videos and evaluates Naïve Bayes, Support Vector Machine (SVM), and Random Forest classifiers against human-validated labels. A vocational higher-education lecturer and an Indonesian-language lecturer annotated all comments independently. They initially agreed on 1,958 comments (97.22%; Cohen’s κ = 0.945), and resolved 56 disagreements by consensus. The final labels comprised 1,227 neutral (60.92%), 702 positive (34.86%), and 85 negative comments (4.22%). A stratified 80/20 split was created before any balancing; the 403-comment test set retained the natural distribution. On the training partition, unigram versus unigram, bigram features and no balancing, class weighting, or training-only undersampling were compared using repeated stratified five-fold cross-validation (two repeats). SVM with unigram TF-IDF and balanced class weights achieved the best cross-validated macro-F1 among the three selected model configurations (0.650 ± 0.054). On the untouched test set, SVM obtained 76.67% accuracy and 0.655 macro-F1, and recovered 6 of 17 negative comments (35.29% recall). Random Forest produced slightly higher accuracy (77.17%) but only 5.88% negative recall, while Naïve Bayes detected no negative cases. The results provide a reproducible screening pipeline that may complement institutional accountability mechanisms, but the limited recovery of negative feedback precludes its use as an autonomous early-warning system.

Toward a Distributed Cardiac Auscultation Simulator: Feasibility of MQTT-Synchronized Multi-Node Audio Playback

2026-09-30T21:03:53+00:00

Cardiac auscultation training traditionally relies on passive manikins or direct patient contact, both of which limit exposure to varied heart sound patterns at multiple anatomical points and remain costly to acquire and maintain. This study investigated the feasibility of the Message Queuing Telemetry Transport (MQTT) protocol, combined with Network Time Protocol (NTP) based timestamp synchronization, for coordinating audio playback across distributed low-cost microcontroller nodes as a first step toward a distributed cardiac auscultation simulator. A prototype consisting of two ESP32 microcontrollers, each driving two independent I2S audio channels through MAX98357A amplifiers, was connected to a Mosquitto MQTT broker running on a laptop acting as a temporary central controller. Three experimental scenarios were conducted: baseline synchronized playback on a single channel per node, synchronized playback on a second channel per node, and simultaneous two-channel playback within a single node. Across 25 to 30 valid trials per scenario, median command latency ranged from 44.5 to 124 milliseconds, and median inter-node synchronization error ranged from 243.5 to 369 milliseconds. Simultaneous two-channel playback within one node revealed a median 9.7-second delay between channels, attributed to blocking audio playback in the current firmware. These results demonstrate that MQTT-based synchronization is technically feasible on low-cost hardware, though the achieved accuracy falls short of a ±50 millisecond target, indicating specific firmware optimizations needed to improve synchronization precision.

Smart Aquarium Maintenance System with Solar-Powered Energy Supply

2026-09-30T21:03:46+00:00

Aquarium maintenance needs constant monitoring of water quality and all its equipment to keep fish and other aquatic life healthy. Manual maintenance is inefficient, since it requires routine checks and control of several parameters at the same time. This study designed a smart aquarium maintenance system to monitor water quality, controls equipment automatically, and draws its power from a renewable energy source. The system used an ESP32 S3 microcontroller as its main controller, integrated with pH, water temperature, turbidity, and ultrasonic sensors. A water pump, a relay module, a servo motor, a Peltier-based cooling module, and a heater worked together to keep the aquarium environment within the desired range. Then a solar power plant charged the battery through an MPPT controller. The Android application, named AquaKeep, was chosen to connect to the hardware through Firebase Realtime Database and lets the user monitor and control the system remotely. Testing showed that the temperature, ultrasonic, and pH sensors reach average errors of 1.13%, 1.98%, and 1.70%, all within the 5% threshold set for this study. The automatic feeding system reached a 100% success rate across eight test scenarios. The cooling system lowered the water temperature by an average of 2.9°C every 10 minutes, and the solar power supply kept the system running for 17 hours without sunlight, exceeding the 12-hour target. These results showed that the developed system works accurately, automatically, and sustainably in keeping the aquarium environment stable.