Publication Search

Given the increasing prevalence of mental health challenges in digital work settings, especially among IT remote workers, early detection mechanisms have become critically important. This study aims to improve the prediction accuracy of mental health conditions among IT remote workers by integrating feature engineering techniques within machine learning models. Five algorithms consisting of Random Forest, Logistic Regression, K-Nearest Neighbors, Decision Tree, and Naive Bayes were evaluated. The Random Forest model achieved the best performance, with 83% accuracy, 83% precision, 100% recall, and a 90% F1-score, followed closely by Logistic Regression with 82% accuracy. Nevertheless, the results demonstrate the feasibility of applying machine learning to support the early detection of mental health risks, offering a strong foundation for future research in predictive analytics and the development of intelligent support systems within digital work environments.

This research presents a personalized, context-aware recommender system to suggest Places of Interest (POIs) using a hybrid approach combining Bayesian inference and collaborative filtering. The system explicitly addresses the cold-start problem that new users face and improves recommendation accuracy by considering contextual variables such as user mood, budget, companion, and location. The system collects real-time contextual inputs for new users with no historical data and applies Bayesian inference to generate relevant POI suggestions. As users begin to interact and provide ratings, the system progressively shifts to a collaborative filtering mechanism, leveraging cosine similarity to identify similar users within comparable contexts. The recommender system focuses on three categories of POIs: restaurants, hotels, and landmarks. These locations are retrieved through the Google Maps API, and only mapped locations are considered. The system was implemented on Android devices and evaluated through a user study involving 25 participants from diverse backgrounds, including software developers, IT students, and general users. Evaluation metrics such as normalized Discounted Cumulative Gain (nDCG) and classification accuracy were used to assess recommendation quality. Results demonstrate that the system performs better than traditional methods, with nDCG improvements reaching up to 83 percent. Users reported high satisfaction regarding the recommendations' accuracy, ease of use, and contextual relevance. While the system offers significant improvements, it also has certain limitations. Its dependency on Google Maps data may restrict its scope, and using only four contextual factors limits the system’s adaptability to more complex user preferences. Future enhancements could include additional dynamic contexts such as weather, POI popularity, and time-related trends, as well as integrating more advanced models to increase personalization and flexibility in real-world applications.

The plan to offer birth control to teenagers, outlined in Government Regulation (PP) No. 28 of 2024, has sparked different responses in the public, especially on social media sites like Twitter. This research intends to look into how people feel about this plan by using the Naïve Bayes Classifier technique. Information was gathered from Twitter by using data collection methods with the snscrape tool and the Python coding language. A total of 1,000 tweets related to the topic of the policy were gathered and went through initial processing steps like cleaning, breaking into words, changing cases, and removing common words. The Naïve Bayes Classifier technique was employed to sort the public's feelings into three groups: positive, negative, and neutral. The findings showed that half of the tweets (50%) had a negative view on the policy, while 35% had a positive outlook, and 15% were neutral. The accuracy of the method used was 78%, with a precision of 74%, a recall of 79%, and an F1-score of 76%. The findings from this research offer a summary of how the public feels about the birth control policy for teenagers, which can help the government assess and create policies that better meet the community's needs and worries. Additionally, this research highlights how well the Naïve Bayes Classifier method works for analyzing sentiments on social media, even though there are some challenges when it comes to understanding language subtleties like sarcasm.

This research aims to address the challenges in sales data management and limited market reach faced by the Inegena Village-Owned Enterprise (BUMDes) in North Bajawa District, Ngada Regency, East Nusa Tenggara. The BUMDes produces and sells candlenut oil, a superior local product, but currently uses a manual sales and recording system (B2B and B2C), which leads to fluctuating demand, difficulties in sales data analysis, and decision-making that lacks valid data. To address these issues, a web-based e-commerce system was implemented. This system was designed using Agile methods, involving planning, implementation, software testing (black box testing), documentation, deployment, and maintenance. Furthermore, the Naïve Bayes algorithm was applied to visualize sales data and support better decision-making by classifying best-selling products, popular payment methods, and sales levels. The results of this research are expected to assist Inegena BUMDes in improving sales efficiency, expanding the market reach of candlenut oil products nationally. This system uses supporting software such as Xampp, PHP, and MySQL.

This study aims to classify the level of satisfaction of Dasawisma cadres with the Carik application in West Semper Village by utilizing the Naive Bayes method. Data was obtained through questionnaires, which were compiled based on three main aspects: ease of use, speed of access, and the usefulness of applications in supporting cadre tasks. After the data is collected, a pre-processing and labeling process is carried out, where the level of satisfaction of respondents is categorized into two classes, namely "satisfied" and "dissatisfied". The Naive Bayes algorithm is applied to predict satisfaction classes based on questionnaire answers. The results of the analysis show that the Naive Bayes method is able to perform classification with sufficient accuracy, so that it can be used as an evaluation tool and decision support in the development of the carik application. This method can also help the management understand user perceptions and improve the system based on objective and routine data in line with the needs of field cadres.

 Election integrity is a key element in ensuring democratic legitimacy, yet election fraud remains a significant threat to this process. This research aims to evaluate the use of Bayesian statistics in detecting anomalies in vote distributions that may indicate potential election fraud. In the context of elections, whose reliability is often questioned, statistical analysis becomes crucial to ensure transparency and credibility in election results. This study uses national election data collected from various polling stations (TPS), focusing on vote distribution per candidate in each TPS. A Bayesian probabilistic model is applied to analyze the observed vote distribution compared to the expected distribution. By using Bayesian inference, researchers can update the probability distribution based on the obtained data, allowing for more accurate anomaly identification. The analysis results show significant deviations in several TPS, where the observed vote distribution exhibits patterns inconsistent with the expected distribution. These anomalies are suspicious as they may indicate manipulation or errors in the vote counting process. The implications of this research suggest that the Bayesian method can be used to improve election transparency and provide a more reliable tool for detecting fraud. The future application of this technique can help increase public trust in election results and support the integrity of the democratic process. This study also recommends further development of the analytical model to expand the scope of anomaly detection and enhance fraud detection accuracy in larger and more complex elections.

Hypertension is a disease that occurs to arteries that causes the supply of oxygen and nutrition that the body needs to be blocked. Hypertension is often called a silent killer, because it is a kind of disease that is very harmful but comes without awareness to its victim. People with hypertension in average are up to 40 years old and it happened all of his after life . In common hypertension caused by heredity, unhealthy lifestyle, and triggered by the more salty consumption, alcohol and stress. An expert system could be the solution to solve the problem because this system works just like an expert and was created by the naïve Bayes method with the rules and basic system that are the same just like the hyperantion desease. Through this application, users can consult with this system just like usually people consult with the expert to diagnose the sign that happened to the user and find the solution of what happened to themselves.

The proliferation of Internet of Things (IoT) devices has introduced significant security challenges, necessitating robust attack detection mechanisms. This study presents a comprehensive comparative analysis of ten supervised learning algorithms for IoT attack detection and classification, addressing the critical challenge of balancing detection accuracy with practical deployment constraints. Using the CICIoT2023 dataset, encompassing data from 105 IoT devices and 33 attack types, we evaluate Naive Bayes, Artificial Neural Networks (ANN), Logistic Regression (LR), k-NN, XGBoost, Random Forest (RF), LightGBM, GRU, LSTM, and CNN algorithms based on some performance metrics. The comparative test results show superior performance to the traditional ensemble approach, with RF achieving 99.29% accuracy and leading precision (82.30%), followed closely by XGBoost with 99.26% accuracy and 79.60% precision. Deep learning approaches also demonstrate strong capabilities, with CNN achieving 98.33% accuracy and 71.18% precision, though these metrics indicate ongoing challenges with class imbalance. The analysis of confusion matrices reveals varying success across different attack types, with some algorithms showing perfect detection rates for certain attacks while struggling with others. The study highlights a crucial distinction in IoT security: while high precision remains important, the potentially catastrophic impact of missed attacks necessitates equal attention to recall metrics, as evidenced by the varying recall rates across algorithms (RF: 72.19%, XGBoost: 71.69%, CNN: 64.72%). These findings provide vital insights for developing balanced, context-aware intrusion detection systems for IoT environments, emphasizing the need to consider performance metrics and practical deployment constraints.

Predicting credit worthiness is an important step for banks to reduce the risk of bad credit. This research compares the performance of four classification algorithms, namely SVM, Naïve Bayes, Random Forest and Decision Tree using simulated datasets. The results obtained on the metrics of accuracy, precision, recall, F1 score, and AUC-ROC, show that Decision Tree has the best performance with 42.5% accuracy, 48.3% precision, 47.5% recall, 47.5% F1 score, and AUC 0.60, indicating its ability to is in differentiating credit worthiness. Random Forest achieved an accuracy of 37.5% and an AUC of 0.493, while Naïve Bayes had the lowest accuracy with an accuracy of 27.5% and an AUC of 0.425. SVM gives better results than Naïve Bayes but is still inferior to Decision Tree. This research recommends implementing a Decision Tree as the main model with optimization through hyperparameter tuning, adding relevant features, and handling data accounting. These results are expected to support banking decision making more effectively and efficiently.

This study aims to classify the level of satisfaction of Dasawisma cadres with the Carik application in West Semper Village by utilizing the Naive Bayes method. Data was obtained through questionnaires, which were compiled based on three main aspects: ease of use, speed of access, and the usefulness of applications in supporting cadre tasks. After the data is collected, a pre-processing and labeling process is carried out, where the level of satisfaction of respondents is categorized into two classes, namely "satisfied" and "dissatisfied". The Naive Bayes algorithm is applied to predict satisfaction classes based on questionnaire answers. The results of the analysis show that the Naive Bayes method is able to perform classification with sufficient accuracy, so that it can be used as an evaluation tool and decision support in the development of the carik application. This method can also help the management understand user perceptions and improve the system based on objective and routine data in line with the needs of field cadres

Gangguan pada tidur adalah suatu kondisi yang ditandai dengan adanya gangguan pada kuantitas, kualitas, dan atau durasi tidur. Gangguan tidur yang umum terjadi adalah insomnia dan sleep apnea. Jika tidak ditangani, insomnia maupun sleep apnea dapat menjadi kronis dan mengakibatkan terkena gejala penyakit yang lebih serius lainnya seperti hipertensi, depresi, penurunan fisik dan juga fungsi fital yang dapat menyebabkan komplikasi penyakit lainnya. Dengan adanya sebuah sistem yang dapat membantu prediksi diagnosis gangguan tidur akan membantu mengetahui gangguan tidur yang diderita dan dapat ditangani dengan segera. Proses prediksi dilakukan menggunakan metode klasifikasi algoritma Naïve Bayes Gaussian. Berdasarkan pengujian data training dan data uji didapatkan akurasi sebesar 85,3%. Hal tersebut membuktikan bahwa klasifikasi menggunakan algoritma Naïve Bayes Gaussian tergolong baik untuk memprediksi gangguan tidur.

Diabetes is one of the diseases that poses a significant global health challenge, with a considerable impact on quality of life and mortality rates. This study examines the use of the Support Vector Machine (SVM) algorithm for diabetes classification through a literature review. SVM was chosen due to its ability to handle imbalanced and complex data. The aim of this study is to assess the effectiveness of SVM compared to other machine learning methods in detecting diabetes. The results of the literature review indicate that SVM achieves higher accuracy than other methods such as Naïve Bayes and Decision Tree, with some studies showing accuracy above 90%. This study is expected to provide deeper insights into the development of machine learning-based diagnostic systems for diabetes.

The 2024 Indonesian Presidential Election marked the fifth general election in the country, aimed at electing a new President and Vice President for the 2024–2029 term. Candidates competed to succeed the outgoing president, who had served two constitutional terms. A key aspect of this election was the candidate debates, where each candidate presented their vision, allowing the public to assess their policies. These debates were broadcast on platforms like YouTube, giving the public a space to comment. However, analyzing YouTube comments presents challenges due to the volume of data, language diversity, and informal expressions. Sentiment analysis, crucial for understanding public opinion, uses algorithms such as Naïve Bayes, which is based on Bayes' Theorem and assumes feature independence. Naïve Bayes is widely used in text analysis for its speed and simplicity. When applied to YouTube comments from the 2024 debates, the algorithm demonstrated its effectiveness, especially with a balanced dataset through random oversampling. It achieved 85.155% accuracy, high precision, recall, and an AUC of 96.8% on an 80:20 data split. Its fast classification time (0.000998 seconds) makes it suitable for real-time sentiment analysis, validating its use for political events. Future applications may incorporate advanced techniques like BERT for more sophisticated analysis.

In today's world of retail and technology, competition is fiercely competitive. With the development of retail businesses increasing in number and mushrooming in a region, consumer needs are increasing, and retail business players are competing to develop their businesses by utilizing existing technology. Daily sales transaction data continues to increase, causing a lot of storage. Toko Ira has more than 228 sales transaction data records from 2023 to 2024 that have not been used. Data requires a lot of storage space. Additionally, the data has not been used in an effective way. Based on this problem, this research aims to use data mining to classify sales transaction data to determine which items are selling best. This research is a case study with a qualitative approach. This research was conducted with the Naive Bayes method and Rapidminer was used. The results of the sales transaction data classification research are the division of products into best-selling and non-selling categories. The results of this research show that the K-Nearest Neighbors (KNN) algorithm with a 50:50 data division is more effective in predicting and classifying sales of best-selling and non-selling products in IRA stores. The results show that the Naive Bayes algorithm has an accuracy of 89.91%, while the K-Nearest Neighbors (KNN) algorithm has an accuracy of 60.09%.

The news about the proposal of the government of the Republic of Indonesia regarding the postponement of the 2024 elections is certainly an interesting discussion. In this research, sentiment analysis will be carried out on the issue of postponing the election. In this study, a dataset obtained using the crawling technique was obtained in the amount of 1280 tweet data about the postponement of the 2024 election. Data labeling in this study uses lexicon-based techniques with Indonesian dictionaries. By applying this technique, the details of the data in the positive class are 67.7%, namely 157 opinion data, and 32.3% negative, namely 75 opinion data. The sentiment classification system's training and test data yield a 9:1 ratio when the Naïve Bayes Classifier method is applied, and word weighting using TF-IDF yields an accuracy value of 91.67%, precision of 90.91%, recall of 100%, and f1-score of 95.24%.

Technological advancements have brought significant transformations across various fields, including the application of machine learning in recommendation and classification systems. Machine learning leverages data processing, utilizes algorithms, and efficiently identifies patterns to produce accurate recommendations and predictions. This study aims to review machine learning-based recommendation system approaches, analyze model performance, and compare the algorithms used. A literature review was conducted by examining journals published in the past five years, focusing on algorithm implementation. The findings indicate that the Naïve Bayes algorithm delivers the best performance, achieving an accuracy of up to 97%. This algorithm is particularly well-suited for processing small to medium-sized datasets with high efficiency. The research provides comprehensive insights into the performance and limitations of various algorithms, serving as a valuable guide for future developments in the field.

Cardiovascular disease is a leading cause of death globally, necessitating effective predictive systems. This research aims to analyze the effectiveness of various machine learning (ML) models—Logistic Regression (LR), Random Forest (RF), Naive Bayes (NB), Support Vector Classifier (SVC), and K-Nearest Neighbors (KNN)—in predicting heart disease using publicly available health data. The study involved pre-processing data, training models, and evaluating them using accuracy, precision, recall, F1-score, and G-Mean metrics. The results show that KNN is the most reliable model, with the highest accuracy of 92%. Significant health features were identified, such as chest pain type and maximum heart rate. The study contributes to improving clinical decision support systems by identifying optimal ML models for heart disease prediction.

The growth of e-commerce in Indonesia has led to the emergence of various online shopping platforms, with Shopee being one of the most popular in Semarang City. User reviews on the Shopee application serve as a valuable data source for analyzing customer satisfaction levels; however, the large volume of data requires a systematic and accurate analytical approach. This study aims to analyze user review sentiments of the Shopee application using three machine learning algorithms: Random Forest, Naïve Bayes, and Support Vector Machine (SVM), as well as comparing the accuracy of these three algorithms. This research utilized 1000 reviews collected through web scraping from the Play Store, which were categorized into three classifications: positive, neutral, and negative sentiments. The analysis process encompassed pre-processing stages, feature extraction using TF-IDF, and classification using Random Forest, Naïve Bayes, and Support Vector Machine algorithms. The results demonstrated that the Random Forest algorithm achieved the highest accuracy at 96.19%, followed by Support Vector Machine with 95.71% accuracy, and Naïve Bayes with 84.76% accuracy. This research highlights the effectiveness of Random Forest and SVM in classifying user review sentiments towards the Shopee application.

Students reading interest is a crucial factor in enhancing the quality of education. However, the lack of structured data makes it challenging to identify specific patterns of reading interest. This study aims to implement a data mining method using the Naive Bayes algorithm to analyze students' reading interest at SMP Negeri 2 Palembang's library. The data used includes book borrowing history, types of books, and library visit frequency over one semester. The analysis results indicate that the Naive Bayes method achieves an accuracy rate of 80% in classifying reading interest based on predetermined categories. These findings are expected to assist the school in designing more effective literacy programs.  

Sentiment Analysis is a technique for extracting text data to obtain information about positive, neutral or negative sentiments. The purpose of sentiment analysis is given by internet users on social media to provide a personal assessment or opinion. Paga Lewu Shop that often gets user sentiment through social media is Paga Lewu Shop. The existence of consumer opinion sentiments about Paga Lewu Shop can be analyzed and utilized to obtain useful information for other customers and the Paga Lewu Shop. By using the Text Mining technique classification method, a sentiment will be known as positive, neutral or negative. One of the algorithms widely used in sentiment analysis is the Naïve Bayes classification method. This study uses the Naïve Bayes Classifier (NBC) method with tf-idf weighting accompanied by the addition of an emotion icon conversion feature (emoticon) to determine the existing sentiment class from tweets about the Paga Lewu Shop. The results of the study show that the Naïve Bayes method without additional features is able to classify sentiment with an accuracy value of 96.44%, while if the tf-idf weighting feature is added along with the conversion of emotion icons, the accuracy value can be increased to 98%.