钓鱼网站数据集

本数据集聚焦钓鱼网站检测，融合网页URL与首页截图，支撑多模态安全模型研发。采集自公开数据与安全渠道，经编码修复、图像质量评估、OCR文本提取及类别平衡重采样等处理，确保数据准确完整，可用于网页风险分析及多模态学习，具高复用价值。

This dataset focuses on phishing website detection, integrating webpage URLs and homepage screenshots to support the development of multimodal security models. Collected from public datasets and security-related channels, it has undergone post-processing steps including encoding correction, image quality assessment, OCR text extraction, and class-balanced resampling, ensuring the accuracy and completeness of the data. This dataset can be applied to webpage risk analysis and multimodal learning, and possesses high reusability.

搜集汇总

背景与挑战

背景概述

钓鱼网站数据集是一个专注于钓鱼网站检测的多模态数据集，融合网页URL和首页截图，适用于网页风险分析和多模态学习。数据集来源于公开数据和安全渠道，经过编码修复、图像质量评估、OCR文本提取及类别平衡重采样等处理，确保数据准确完整，具有高复用价值。