phishrepo-dataset

PhishRepo is implemented to fill the data gap in the anti-phishing domain, and it is still at an experimental level. PhishRepo collects the data available here during its testing stage, and the dataset includes verified phishing webpages. The provided dataset contains diverse information sources from the latest phishing pages. The diverse feature-rich data present in the dataset is a current need in the machine learning-based anti-phishing domain to overcome inept learning models in phishing detection. The dataset can be used to analyse significant phishing features, experiment with different feature extraction techniques, effectively try out some representation learning techniques such as deep learning from these raw data at a practical level. The dataset contains an index.csv file, and it will be the main file that should be used when mapping index file content with available folders.PhishRepo 旨在填补反钓鱼领域的数据空白，目前仍处于实验阶段。该数据集在测试阶段收集了此处可用的数据，包含经过验证的钓鱼网页。本数据集涵盖源自最新钓鱼网页的多样化信息源，其蕴含的多样化且富含特征的数据，正是当前基于机器学习的反钓鱼领域所需，以解决钓鱼检测中学习模型效能不佳的问题。该数据集可用于分析关键钓鱼特征、测试多种特征提取技术，还能在实际场景中基于这些原始数据有效尝试表征学习技术（如深度学习）。该数据集包含一个index.csv文件，该文件是将索引文件内容与可用文件夹进行映射时的核心使用文件。