Journal of Informatics And Telecommunication Engineering
Not a member yet
373 research outputs found
Sort by
Grouping of Tourism Locations in Indonesia Using Distance Variations in the K-Means Algorithm
Full text linkIndonesia is home to a diverse range of tourist destinations, yet the classification and mapping of these locations remain a challenge in tourism management. This study aims to cluster tourist destinations in Indonesia by applying the K-Means algorithm with three distance metric variations: Euclidean Distance, Manhattan Distance, and Canberra Distance. The dataset was sourced from public data repositories and underwent preprocessing steps, including data normalization. The optimal number of clusters was determined using the Elbow Method, while the clustering results were evaluated using the Silhouette Score and Davies-Bouldin Index. The findings indicate that Manhattan Distance produced the highest Silhouette Score (0.321463), suggesting superior clustering performance compared to the other two metrics. The results of this study provide valuable insights for stakeholders in formulating strategic tourism promotion and infrastructure development efforts
Enchancing Brain Tumor Disease Classification via SqueezeNet Architecture Integrated with Group Convolution
Full text linkBrain tumor classification using MRI images is a major challenge in medical image processing, particularly when facing imbalanced data between classes. This imbalance often leads to model bias toward the majority class and reduces sensitivity to the minority class—patients with tumors. This study aims to analyze the impact of applying Group Convolution techniques to the VGG19 and SqueezeNet architectures to enhance both computational efficiency and classification accuracy. A quantitative experimental approach was employed, implementing Convolutional Neural Networks (CNNs) using the PyTorch framework. The dataset includes two classes, “Yes” (with tumor) and “No” (without tumor), organized into Train, Validation, and Test folders. The models were evaluated by comparing the performance of standard architectures with modified versions integrating Group Convolution. Experimental results show that SqueezeNet with Group Convolution achieved up to 90% accuracy, outperforming the original model. Additionally, the model exhibited significantly improved sensitivity to the minority class, indicating better performance under imbalanced conditions. These findings suggest that Group Convolution enhances not only computational efficiency but also classification capability. Therefore, this technique is applicable in developing automated diagnostic systems. Future research is encouraged to combine Group Convolution with methods such as attention mechanisms to achieve more optimal and reliable classification results
Sentiment Analysis of Public Opinion on Online Gambling Through Social Media Using Convolutional Neural Network
Full text linkOnline gambling has become a serious social issue due to its easy accessibility through digital platforms, requiring effective policy interventions. This study analyzes public sentiment toward online gambling by examining 10,000 YouTube comments using a Convolutional Neural Network (CNN) algorithm. Data were collected via the YouTube API and underwent preprocessing steps including text cleaning, normalization, tokenization, stopword removal, and stemming. Sentiment labeling was performed using a lexicon-based approach, with data transformed through Word2Vec embedding and balanced using oversampling techniques. The CNN model, consisting of embedding, convolutional, pooling, and dense layers, achieved an impressive accuracy of 99.10%, outperforming traditional machine learning methods. Sentiment was categorized into positive, neutral, and negative, with the majority of comments reflecting positive sentiment, indicating public support for efforts to combat online gambling. WordCloud visualizations highlighted dominant themes and frequently used terms. This study demonstrates the effectiveness of CNN in analyzing unstructured social media data and offers valuable insights for policymakers. Future research should explore hybrid architectures such as CNN-LSTM and expand datasets by including other platforms like Twitter, Instagram, and TikTok to enhance generalization and address broader social challenges
CatBoost Algorithm Implementation for Classifying Women's Fashion Products
Full text linkThe rapid growth of the women's fashion industry in the digital era has intensified the need for data-driven approaches to understand customer preferences. This study aims to classify women’s clothing products based on customer reviews by applying CatBoost, a gradient boosting algorithm known for its strong performance with categorical features. The dataset, consisting of 23,486 entries and 11 attributes, was obtained from Kaggle and processed through data cleaning, normalization, exploratory analysis, and model training. Hyperparameter optimization was conducted using Grid Search. Model performance was evaluated using accuracy, precision, recall, and F1-score, and benchmarked against four traditional classifiers: Decision Tree (C4.5), Naïve Bayes, Support Vector Machine (SVM), and K-Nearest Neighbor (KNN). The results show that CatBoost achieved an accuracy of 93.70%, an F1-score of 0.9606, and an AUC of 0.9691, indicating excellent and balanced classification performance. This study demonstrates the effectiveness of CatBoost in handling customer review data and contributes to the development of intelligent product classification systems in the fashion industr
Risk Analysis In Indonesian Educational Online Learning Systems: A Systematic Literature Review
Full text linkOver the past few years, the idea of online learning has gained popularity and due to the COVID-19 pandemic, has become essential everywhere, including in Indonesia. Nevertheless, the deployment of e-learning management systems has brought about several IT hazards that affect academic operations' user experience security and overall operational efficacy worldwide. This study uses a systematic literature review (SLR) of research publications indexed in Google Scholar to examine the possible dangers related to e-learning management systems in academic institutions. The classification, assessment, and identification of the hazards that educational institutions encounter while incorporating e-learning systems into their infrastructure are the main objectives of this study. Issues like operational failures, obsolete hardware and software, cybersecurity threats, network accessibility and stability concerns, data privacy, and illegal access are the main topics of this study. Additionally, this study highlights the necessity of more effective and focused risk mitigation techniques created especially to meet the demands of Indonesian academic setting
Development of a Prototype System for Monitoring and Controlling Cat Care Based on the Internet of Things
Full text linkIn today's fast-paced and stressful daily life, many people experience fatigue and stress, necessitating solutions to alleviate these feelings. Taking care of pets, especially cats, can have a positive impact on human mental health. However, caring for cats requires time and effort, particularly in monitoring their health and the environmental conditions of their living space. This study aims to design and develop a prototype of a cat care monitoring and control system based on the Internet of Things (IoT), enabling real-time monitoring and automated control of various aspects, including cat health, body temperature, and feeding. The system integrates multiple components, including ESP-32, MLX90614 GY-906 Sensor, Ultrasonic Sensor, DHT11 Sensor, RTC DS1302, Servo Motor, Relay, LCD 16x2, 5V Water Pump, Exhaust Fan, and the Telegram application as a monitoring interface. The study's findings indicate that air-conditioned rooms provide the most stable and comfortable temperature and humidity conditions for both pets and humans. Additionally, the DHT11 sensor is more suitable for indoor humidity measurement, whereas the DTH sensor is more optimal for outdoor environments. The ultrasonic sensor proves to be practical for distance measurement, while the load cell delivers precise results for weight measurement requiring high accuracy and stability. This research offers an innovative solution to enhance cat monitoring and care, contributing to an overall improvement in pet quality of life
Analysis of Recommendation System on Travel Platform Using Content-Based Filtering and Collaborative Filtering Algorithms at PT. Angkasa Tour & Travel
Full text linkThis study aims to evaluate the effectiveness of the recommendation system on the PT Angkasa Tour & Travel travel platform using content-based filtering and collaborative filtering algorithms. The background of the identified problem is the need to improve the accuracy and relevance of recommendations in the travel platform, which functions to assist users in choosing travel services that suit their preferences. This research method includes an analysis of the application of the content-based filtering algorithm that focuses on the characteristics of individual users and products, as well as the collaborative filtering algorithm that utilizes collective user behavior patterns. The results of the study indicate that content-based filtering is effective in providing recommendations based on specific user preferences and product attributes, while collaborative filtering is able to produce recommendations based on collective user behavior patterns. This study also reveals that the combination of the two approaches can improve the accuracy and relevance of recommendations, thus better meeting user needs. The conclusion of this study is that the integration of content-based and collaborative filtering in the recommendation system can provide a more comprehensive solution to meet user preferences and needs on the PT Angkasa Tour & Travel travel platform
Web-Based Job Portal System with WhatsApp Integration for Interview Invitation and Verification Automation
Full text linkIn the digital era, job searches increasingly rely on online platforms that provide real-time job vacancy information. However, the conventional recruitment process still faces various obstacles, such as delays in information to applicants, the amount of administrative costs used by applicants in preparing documents to apply for jobs, vulnerable to fraudulent locker information from fake companies which of course this is detrimental and has an impact on job applicants. WhatsApp as the most popular communication media in Indonesia is a strategic opportunity in answering these challenges. This research aims to develop a digital-based job vacancy application with an integrated WhatsApp-based automatic verification interview invitation system and offer applicants the convenience of applying for a job and the ease of the Company in managing job applicant data to be more effective and efficient. This study is on the Instagram account @infokerjambi job vacancy media in Jambi Province with 50.1 thousand followers and 1.2 million profile impressions. Based on data from BPS Jambi Province, the Open Unemployment Rate (TPT) is 4.45. The Waterfaal model method used in this research and the test results show that the system is able to speed up the information process of sending interview invitations, reduce the potential for information delays because the system presents interview invitation features to wa applicants, applications and applicant emails in real time and increase security and trust in the recruitment process. For applicants, this system provides easy access to valid and real-time interview information, thus reducing the risk of fraud from verified companies. For companies, this system improves the management of applicant data management and administrative efficiency to reach a wider range of candidates
Reliability Analysis of the Circulating Water Pump Instrumentation System Using the FMEA Method at PT PLN Nusantara Power UP Tenayan
Full text linkPT PLN Nusantara Power UP Tenayan in Pekanbaru operates a power plant that relies on the Circulating Water Pump (CWP) as a vital part of the cooling system. Based on the results of field observations and interviews, failures in the CWP instrumentation system can cause downtime and reduce operating efficiency, but reliability studies are still limited. This study aims to analyze the reliability of the CWP instrumentation system using the Failure Mode and Effect Analysis (FMEA) method. Data were obtained through field observations and technician interviews, then analyzed based on Severity, Occurrence, and Detection parameters. The analysis identified eight main components, with Risk Priority Number (RPN) values all below the 200 threshold. Based on the results of the FMEA calculation, the limit switch component has the highest RPN value of 160 with the potential for downtime reaching 2 to 3 hours per occurrence. The application of FMEA is proven effective to reduce the risk of failure by 25% based on estimated technical evaluation and failure history
Tourist Classification Based on Consumer Behavior Using XGBoost Algorithm
Full text linkThis study discusses the application of the XGBosst Algorithm to Tourists based on consumer behavior. The purpose of this study is to predict or analyze tourist review data, and to help provide and understand needs so as to improve the quality of services offered. Indonesia has great tourism potential thanks to its natural beauty and cultural diversity. This sector plays an important role in the national economy by creating jobs and encouraging the creative industry and hospitality. The presence of tourists increases regional income through taxes and spending in sectors such as hotels, restaurants, and souvenir shops, as well as creating new jobs. In addition to tourists being able to increase income, there is a need for an understanding of each tourist behavior that is important for the development of adaptive and sustainable tourism