To address issues such as low testing efficiency and quality assurance challenges in the rapid iterative development model of Web applications, Research designs and develops a new automated testing framework based on Selenium automation testing tools, employing a data-driven approach and the Page Object (PO) framework concept. The framework achieves separation of test data, page elements, and scripts through layered design, enhancing script reusability and maintainability, and supports Automatic Unattended Testing. For technical challenges such as captcha recognition and special element positioning, solutions based on Optical Character Recognition (OCR) and image matching positioning technology are proposed, improving the adaptability and scalability of the framework. Research results show that the framework can significantly reduce manpower costs, improve testing efficiency, and effectively ensure software quality during frequent iterative deliveries.