Trực quan hóa dữ liệu gói cơ bản và nâng cao

Analysis of funding requests and the degree of European funds absorption

Adopted from learning course textbook.

Breast cancer remains one of the most significant global health challenges, where early and accurate diagnosis plays a critical role in improving patient survival outcomes. In this project, I explore a machine learning approach to breast cancer classification using the Wisconsin Diagnostic Breast Cancer dataset, which contains quantitative morphological features extracted from digitized images of fine needle aspirates (FNA). The workflow covers the full data science pipeline, including data cleaning, feature standardization, exploratory data analysis, and correlation structure investigation. Key tumor characteristics such as radius, perimeter, area, concavity, and compactness were analyzed to understand their relationship with diagnosis outcomes. A K-Nearest Neighbors (KNN) classifier was developed to distinguish between benign and malignant tumors based on these morphological features. The model achieved high predictive performance, demonstrating strong sensitivity and specificity, and highlighting the effectiveness of simple distance-based learning methods in biomedical classification problems when combined with proper preprocessing. This project demonstrates how statistical learning techniques can support clinical decision-making by uncovering meaningful patterns in biomedical data and contributing to early cancer detection research.

Analisis Structural Equation Modeling (SEM) untuk mengukur dampak standarisasi terhadap kinerja perusahaan. Data: Survey Standarisasi Bappenas (n=276). Metode: CFA + SEM dengan estimator MLR, Bootstrap 1000x untuk mediasi. Penulis: Yuli & Arifin | Universitas Airlangga × Bappenas | 2025.

LB2