Thyroid cancer is the most common endocrine malignancy, with a steadily increasing incidence rate. Although the overall survival rate is relatively high, the risk of recurrence after definitive treatment such as Radioactive Iodine (RAI) therapy remains a significant clinical challenge. Predicting recurrence risk is crucial for optimizing monitoring strategies and interventions. With advances in technology, machine learning (ML) approaches are increasingly utilized to support medical predictions, including the recurrence of thyroid cancer. This study aims to evaluate the performance of four classification algorithms—Logistic Regression, XGBClassifier, Random Forest Classifier, and Voting Classifier—in predicting thyroid cancer recurrence using the Thyroid Cancer Recurrence After RAI Therapy dataset, which consists of 383 patient records and 13 key clinical attributes. The evaluation was conducted using accuracy, precision, recall, F1-score, and area under the curve (AUC) metrics. The results show that the XGBClassifier is the best-performing model with an accuracy of 97.4% and an AUC of 0.95, demonstrating superior performance in handling the minority class. This research is expected to contribute to the development of more effective machine learning–based clinical decision support systems for predicting thyroid cancer recurrence after therapy.

Kegagalan jantung (Heart Failure) merupakan salah satu penyebab utama kematian dan morbiditas secara global, dengan lebih dari 26 juta orang terdampak di seluruh dunia. Deteksi dini kegagalan jantung sangat penting untuk mencegah komplikasi yang lebih serius, namun diagnosis sering kali sulit dilakukan karena gejala yang tidak spesifik. Pembelajaran mesin (machine learning) telah banyak digunakan untuk membantu prediksi risiko kegagalan jantung melalui analisis data medis. Berbagai algoritma klasifikasi telah diimplementasikan dalam berbagai penelitian sebalumnya, namun belum ada konsensus mengenai algoritma terbaik. Penelitian ini bertujuan untuk melakukan komparasi performa beberapa algoritma klasifikasi dalam memprediksi gagal jantung, menggunakan dataset Heart Failure Prediction dari Kaggle. Algoritma yang diuji termasuk Logistic Regression, K-Nearest Neighbors (KNN), Decision Tree, Random Forest, GaussianNB, dan LightGBM. Hasil menunjukkan bahwa algoritma Random Forest memberikan performa terbaik dengan akurasi 95 persen, sedangkan algoritma KNN menunjukkan akurasi terendah sebesar 76 persen. Penelitian ini diharapkan dapat berkontribusi dalam pengembangan sistem pendukung keputusan klinis yang lebih baik untuk prediksi kegagalan jantung.
 
Kata Kunci: Heart Failure, mesin pembelajaran algoritma klasifikasi

Chicken and duck eggs are highly nutritious food products that are popular among the Indonesian community. However, their perishable nature leads to a decline in quality if not further processed. The salting process is a simple preservation method that can also enhance the flavor of salted egg products. To produce salted egg products with high organoleptic quality preference, a combination of Analytical Hierarchy Process (AHP) and Simple Additive Weighting (SAW) is implemented as a quality decision-making method based on: (1) the experts viewpoints on the importance level of organoleptic criteria, and (2) the scores of organoleptic tests for the criteria of texture, taste, and appearance. In this study, chicken and duck eggs were processed into salted eggs using a salting mixture made of a blend of scouring ash and table salt with a mass ratio of 5:4 and 5:6, resulting in product combinations coded as CE-5/4, CE-5/6, DE-5/4, and DE-5/6. Data processing results on the importance level from the experts viewpoints using the AHP method resulted that the criteria for texture, taste, and aroma had final priority weights of 0.244, 0.617, and 0.139, respectively. Final decision-making using the SAW method indicated that the highest to lowest final preference scores were obtained by products CE-5/4, DE-5/4, DE-5/6, and CE-5/6, with the final preference scores of 0.979, 0.963, 0.931, and 0.906, respectively. The results demonstrated that AHP and SAW were successfully implemented to assist in making decisions regarding the quality of salted egg products based on their organoleptic characteristics.
 
Keywords: salted egg, organoleptic, decision-making, AHP, SAW

In a train ticket booking also has an obstacle, to overcome the obstacles found in conventional ticket booking such as piling up queues at the station, limited time owned by consumers, the process of delivering information about ticket prices for the day of departure or the previous date that cannot be processed, then PT Kereta Api (Persero) Palembang provides solutions through applications that will be made to facilitate ticket booking, Fast transaction process, delivery of information about schedules and ticket prices that can be accessed quickly. In designing this system using a method, namely the Prototype method.

Serangan jantung adalah salah satu faktor utama dalam tingginya tingkat angkat penyebab kematian di seluruh dunia dan memerlukan prosedur diagnosa yang canggih sehingga dapat mengakibatkan peningkatan biaya yang signifikan. Memprediksi penyakit jantung merupakan tantangan utama dalam bidang kesehatan karena keterbatasan peralatan diagnosis penyakit ini. Prediksi penyakit jantung yang akurat sangat penting untuk mengobati pasien sebelum serangan jantung terjadi. Prediksi ini dapat dicapai dengan menggunakan model pembelajaran mesin (machine learning) yang optimal dengan data layanan kesehatan yang kaya (datasets) mengenai penyakit jantung. Namun, Permasalahan yang umumnya dihadapi oleh model prediksi penyakit jantung seperti data yang menyimpang secara ekstrim (outliers), data yang hilang, data yang tidak konsisten, dan data yang tercampur baik secara numerik maupun kategorikal. Data yang tidak konsisten menyebabkan kemungkinan kesalahan prediksi dan akan mempengaruhi hasil dari prediksi. Pada penelitian ini kami mencoba mengatasi masalah outlier pada dataset penyakit jantung menggunakan salah satu metode feature scaling yaitu robust scaler. Hasil Eksperimen dengan model klasifikasi algoritma K-Nearest Neighbors menggunakan metode scaling robust scaler memperoleh nilai lebih baik dibandingkan dengan tanpa robust scaler dengan nilai F1 score sebesar 0.86.