11 research outputs found
Optimasi Parameter Machine Learning untuk Prediksi Penyakit Parkinson dengan Seleksi Fitur dan Teknik Synthetic Minority Over-Sampling
No full textPenyakit Parkinson adalah penyakit neurodegeneratif yang mempengaruhi gerakan tubuh sehingga membutuhkan diagnosis dini untuk penanganan yang tepat. Gejala utama pada penyakit ini meliputi tremor, kekakuan otot, gerakan yang lambat, dan kesulitan mengontrol gerakan. Penyakit Parkinson yang berhasil dideteksi lebih awal dapat diberikan tindakan medis yang efektif. Penelitian terbaru telah menunjukkan bahwa pemrosesan data dengan machine learning dapat menjadi indikator yang berguna untuk deteksi dini penyakit Parkinson. Permodelan prediksi penyakit Parkinson yang akurat masih sulit disebabkan oleh ketidakseimbangan data antara penderita dan orang sehat. Penggunaan seleksi fitur, teknik untuk mengatasi ketidakseimbangan data, dan optimasi pada metode machine learning diharapkan dapat meningkatkan akurasi prediksi penyakit Parkinson secara signifikan. Pada penelitian ini dilakukan prediksi penyakit Parkinson dengan menggunakan dataset yang diambil dari Kaggle.com. Dataset ini memiliki total data 756 record di mana 192 adalah orang sehat dan 564 adalah orang yang menderita penyakit Parkinson. Berdasarkan hasil penelitian, seleksi fitur menggunakan Information Gain dan Gain Ratio terbukti efektif dalam meningkatkan akurasi model sebesar 1,33% dengan penggunaan 4% fitur teratas sehingga menghasilkan akurasi tertinggi pada model CatBoost. Selain itu, model XGBoost dan Random Forest dengan akurasi masing-masing sebesar 91,39%. Teknik penyeimbangan data menggunakan SMOTE berhasil mengatasi ketidakseimbangan data dan meningkatkan kinerja model prediksi. Hal tersebut ditunjukkan dengan model CatBoost, XGBoost, dan Random Forest mencapai akurasi yang sama sebesar 92,05% setelah data diseimbangkan. Selanjutnya, optimasi parameter melalui GridSearchCV menunjukkan peningkatan performa yang signifikan masing-masing sebesar 4,64% pada model CatBoost dan 2,64% pada XGBoost setelah proses optimasi.
===================================================================================================================================
Parkinson's disease is a neurodegenerative condition that affects body movement, necessitating early diagnosis for proper management. Major symptoms of this disease include tremors, muscle stiffness, slow movement, and difficulty controlling movements. Early detection of Parkinson's disease enables effective medical intervention. Recent research has shown that data processing with machine learning can serve as a useful indicator for early Parkinson's disease detection. Accurate Parkinson's disease detection remains challenging due to data imbalance between Parkinson's and non-Parkinson's individuals. The use of feature selection, techniques to address data imbalance, and optimization in machine learning methods are expected to significantly improve Parkinson's disease prediction accuracy. In this study, Parkinson's disease prediction was conducted using a dataset obtained from Kaggle.com. The dataset comprises a total of 756 records, of which 192 represent healthy individuals and 564 represent individuals with Parkinson's disease. Based on the research findings, feature selection using Information Gain and Gain Ratio proved effective in improving model accuracy by 1.33% when utilizing the top 4% of features, resulting in the highest accuracy achieved by the CatBoost model. Additionally, the XGBoost and Random Forest models achieved respective accuracies of 91.39%. The data balancing technique using SMOTE successfully addressed data imbalance and enhanced the performance of prediction models. This was demonstrated by the CatBoost, XGBoost, and Random Forest models achieving the same accuracy of 92.05% after data balancing. Furthermore, parameter optimization through GridSearchCV showed significant performance improvements, with increases of 4.64% in the CatBoost model and 2.64% in the XGBoost model following the optimization process
Analisis Optimalisasi Model Sistem Akuntansi Biaya Bahan Baku Untuk Meningkatkan Efisiensi Operasional di PT. Pancakarsa Bangun Reksa
No full textCost is an economic sacrifice by companies that generate profits in the future, but companies must still strive to make cost efficiency so as not to waste costs. Cost efficiency is a measure of success that is judged by the amount of resources sacrificed to obtain a particular result. The purpose of the study is to determine whether optimizing raw material costs can improve the Company\u27s operational efficiency. Based on this, the author is encouraged to try to describe and explain how the level of raw material efficiency towards the optimization of the raw material cost system at PT. Pancakarsa Build Mutual Funds. In this study, the formulation of the problem that the author raised was. Whether the level of optimization of the raw material cost accounting system model applied by the Company is efficient in its operations. This research was conducted at PT. Pancakarsa Build Mutual Funds. The method used in this study is the descriptive method. The data source used is secondary data
Curing Cholera: Pathogens, Places and Poverty in South Asia
No full textIn this paper I will seek to provide a new understanding of endemicity of disease in India. Through a study of cholera research in the twentieth century I will argue that disease and its endemicity has to be understood in biological factors as well as within a wider social and economic context. I will discuss the medical efforts at locating the causality of cholera from the nineteenth century in Indian climate, water bodies and human anatomy to show that cholera is no more a biological phenomena than water is an ecological or environmental problem. Both are essentially political and economic questions
Manajemen Kualitas Data Dalam Aplikasi E-Commerce: Memastikan Informasi Produk yang Akurat
No full textThis article uses a qualitative methodology with literature review or desk research. Literature research is an approach that observes and original explanations that already exist in relevant literature. In this research, technical techniques are used to collect data, where the author collects data quality management information. Data quality management in e-commerce applications is very important to ensure accurate and relevant product information. Quality data should not have data quality issues such as duplicate, incomplete, inconsistent, incorrect, unclear, disorganized, and insecure data. In e-commerce applications, data quality management is essential to ensure the product information presented to consumers is accurate and meaningful. Inaccurate product information can undermine consumer trust and customer satisfaction. Therefore, e-commerce applications must have an effective information quality management system to ensure accurate and up-to-date product information
양성평등의 일면 : 나그함마디 도마문서에 나타난 살로메 전승
No full textIf one regards the context of the Christian cannonical texts, each author simply delivers the names of some early Christian women, providing insignificant information about them. The first Gospel(Matthew) presents the female followers of Mary Magdalene, Mary the mother of James and Joses, and the mother of Zebedees sons(27:55). The Markan text lists three female disciples : Mary Magdalene, Mary the mother of James the younger and of Joses, and Salome(15:40, 47, and 16:1), as well as many other women(16:4). The female characters are briefly seen as dydwitnesses to the death, the burial and the empty tomb of Jesus.이 논문은 고대 이집트 콥트 영지주의 기독교사에 나타난 도마 공동체의 살로메라는 인물의 종교사회적인 위상을 해석한 것이다. 성서 저자들은 다수의 여성인물들을 거론하지만 그들의 정체에 대해서는 남성 중심적 시각에서 폄하하는 경향이 있다. 이런 부정적인 시나리오에 따르면, 여성들은 기독교 창시자를 추종하는 무리의 일부에 불과할 뿐 아니라 성서 이야기에서도 주요 인물이 아니었기 때문에 여성의 사도성이나 지도자적 성격이 인정받지 못한다. 살로메에 대한 기존의 평가 역시 이런 반여성적인 경향과 무관하지 않다. 그렇다면 이런 불평등한 성차(gender)의 맥락이 다섯 번째 복음서라고 불리는 도마문서에서도 변함없이 전승되고 있을까? 만약 도마문서 내의 살로메(Thomasine Salome)가 단순히 구경꾼 무리에 속한 것이 아니었다면 도마 공동체 구성원들에게 이 인물은 어떻게 알려져 있었을까? 초기 유대기독교의 반여성적 이데올로기가 사히딕 콥트(Sahidic Coptic) 문서의 역사적 가치를 재조명할 때 어떤 종교적 함의를 가질 수 있을까
Pendekatan Metode Ensemble Learning untuk Prakiraan Cuaca menggunakan Soft Voting Classifier
No full textWeather conditions are one of the crucial factors that need attention. Changes in weather conditions significantly impact various activities. Weather condition changes are determined by numerous factors, often occurring within a relatively short period in the atmosphere, such as pressure, wind speed, rainfall, temperature, and other atmospheric phenomena. Issues in weather forecasting arise due to several factors, namely the fluctuating atmospheric conditions. This research proposes the development of a weather forecasting model using the ensemble learning method approach. The weather data used consist of 33746 records with attributes used after preprocessing, namely Temperature, Dew Point, Humidity, Wind Speed, Wind Gust, Pressure, Precipitation, and Condition. Testing in this research employs several single-machine learning methods such as K-Nearest Neighbor (KNN), Logistic Regression, Random Forest, Naive Bayes, and Multi-Layer Perceptron. The Naive Bayes method using default parameters achieves a high accuracy of 99.00%. In the ensemble method, combinations of three methods exhibit excellent accuracy for all combinations. The best combination methods are found in the Soft Voting Classifier method (Random Forest, MLP, Naive Bayes), Soft Voting Classifier (Logistic Regression, MLP, Naive Bayes), and Soft Voting Classifier (Random Forest, KNN, Naive Bayes) with an accuracy of 99.03%.
 
Klasifikasi Pemohon Pinjaman dengan Hyperparameter Tuning dan Teknik Penyeimbangan Data
No full textLoan classification is a critical component of credit risk management, as it categorizes loans based on risk levels and supports the financial stability of banks, where loan-related income represents a substantial share of assets. Effective classification aims to ensure secure asset allocation, minimize credit risk, and prevent potential repayment issues. This study enhances loan classification performance through two strategies: hyperparameter optimization of Decision Tree and Random Forest algorithms, and data balancing techniques to address class imbalance. Experimental results show that the Decision Tree achieves 89.21% accuracy with an F1-Score of 70.17%, while the Random Forest demonstrates higher performance, reaching 94.04% accuracy and an F1-Score of 79.75%. Random Oversampling reduces bias toward majority classes by improving model sensitivity, while hyperparameter tuning with GridSearchCV identifies optimal parameter settings, thereby strengthening predictive performance. The findings highlight that combining data balancing with hyperparameter optimization effectively improves accuracy and F1-Scores. These approaches are not limited to the algorithms tested but can also be applied to other classification methods, offering broader potential for enhancing credit risk prediction in banking
Determinants for Bullying Victimization among 11–16-Year-Olds in 15 Low- and Middle-Income Countries:\ud A Multi-Level Study
No full textBullying is an issue of public health importance among adolescents worldwide. The present study aimed at explaining differences in bullying rates among adolescents in 15 low- and middle-income countries using globally comparable indicators of social and economic well-being. Using data derived from the Global School-based Health Survey, we performed bivariate analyses to examine differences in bullying rates by country and by bullying type. We then constructed a multi-level model using four fixed variables (age, gender, hunger and truancy) at the individual level, random effects at the classroom and\ud
school levels and four fixed variables at the country level (Gini coefficient, per capita Gross Domestic Project, homicide rate and pupil to teacher ratio). Bullying rates differed significantly by classroom, school and by country, with Egypt (34.2%) and Macedonia (3.6%) having the highest and lowest rates, respectively. Eleven-year-olds were the most likely of the studied age groups to report being bullied, as was being a male. Hunger and truancy were found to significantly predict higher rates of bullying. None of the explanatory variables at the country level remained in the final model. While self-reported bullying varied significantly between countries, the variance between classrooms better explained these differences. Our findings suggest that classroom settings should be considered when designing approaches aimed at bullying prevention.\u
Skill Flow: A Fundamental Reconsideration of Skilled-Worker Mobility and Development
No full textLarge numbers of doctors, engineers, and other skilled workers from developing counties choose to move to other countries. Do their choices threaten development? The answer appears so obvious that their movement is most commonly known by the pejorative term “brain drain”. This paper reconsiders the question starting from the most mainstream, explicit definitions of “development”. Under these definitions, it is only possible to advance development by regulating skilled workers’ choices if that regulation greatly expands the substantive freedoms of others to meet their basic needs and live the lives they wish. Much existing evidence and some new evidence suggests that regulating skilled-worker mobility itself does nothing to address the underlying causes of skilled migrants’ choices, generally brings few benefits to others, and instead brings diverse unintended harm. The paper concludes with examples of effective ways that developing countries can build a skill base for development without regulating human movement. The mental shift required to take these policies seriously would be aided by dropping the sententious term “brain drain” in favor of the neutral, accurate, and concise term “skill flow”.skill, talent, professional, educated, graduate, degree, labor, global
Skill Flow: A Fundamental Reconsideration of Skilled-Worker Mobility and Development
No full textLarge numbers of doctors, engineers, and other skilled workers from developing countries choose to move to other countries. Do their choices threaten development? The answer appears so obvious that their movement is most commonly known by the pejorative term “brain drain.” This paper reconsiders the question, starting from the most mainstream, explicit definitions of “development.” Under these definitions, it is only possible to advance development by regulating skilled workers’ choices if that regulation greatly expands the substantive freedoms of others to meet their basic needs and live the lives they wish. Much existing evidence and some new evidence suggests that regulating skilled-worker mobility itself does little to address the underlying causes of skilled migrants’ choices, generally brings few benefits to others, and often brings diverse unintended harm. The paper concludes with examples of effective ways that developing countries can build a skill base for development without regulating human movement. The mental shift required to take these policies seriously would be aided by dropping the sententious term “brain drain” in favor of the neutral, accurate, and concise term “skill flow.”brain drain; migration; development; labor; education; developing; labor mobility; circular migration; higher education; university; training; skilled; high skill; talent; globalization; health workers; high tech; technology transfer
