Scholarlink

This paper presents a machine learning model for address selection. The model aims to improve delivery success rates for special mail services. It integrates multiple data types including customer profiles, geographic features, and historical behavior patterns. This integration enables intelligent prediction of optimal delivery addresses.The model combines Logistic Regression and XGBoost algorithms. This hybrid approach processes multi-dimensional features and generates credibility scores. Experimental results demonstrate strong performance: over 94% accuracy, 0.9 AUC, and 0.46 KS value on validation data. These metrics indicate high stability and prediction precision. We conducted feature importance analysis and ROC validation. These tests confirm the model’s effectiveness in address selection. These research findings provide reliable technical support and a theoretical basis for many scenarios. Such as financial services, legal document delivery, and public governance.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Construction and Application of a Machine Learning-Based Address Selection Model

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Shuai Yue,
  • Fangyu Chen,
  • Lihua Wang,
  • Hongbo Yu

摘要

This paper presents a machine learning model for address selection. The model aims to improve delivery success rates for special mail services. It integrates multiple data types including customer profiles, geographic features, and historical behavior patterns. This integration enables intelligent prediction of optimal delivery addresses.The model combines Logistic Regression and XGBoost algorithms. This hybrid approach processes multi-dimensional features and generates credibility scores. Experimental results demonstrate strong performance: over 94% accuracy, 0.9 AUC, and 0.46 KS value on validation data. These metrics indicate high stability and prediction precision. We conducted feature importance analysis and ROC validation. These tests confirm the model’s effectiveness in address selection. These research findings provide reliable technical support and a theoretical basis for many scenarios. Such as financial services, legal document delivery, and public governance.