As a convenient information collection and feedback tool, the network suggestion box plays an important role in improving the quality of university education and management efficiency. However, the traditional network suggestion box has many shortcomings in access mode, feedback efficiency, information processing and so on. In recent years, with the rise of WeChat mini programs, it has provided a more efficient and convenient channel for teachers and students to give feedback. Combining the convenient development advantages of WeChat mini program platform and the powerful processing power of text analysis algorithm, this paper designs and implements a college network suggestion box based on WeChat mini program. The system is divided into user module for opinion submission, university management module for opinion collection and feedback, and system management module. The development of the system adopts front-end and back-end separation mode. The user module is implemented by WeChat small program development framework, and the university management and system management module are developed by Vue.js framework. The backend uses the Java language, and adopts the popular SpringBoot framework for development. In the processing of the opinion content, the TextRank keyword extraction algorithm is used to analyze the data of the opinion content, and the function of keyword extraction and automatic summary is realized. The content of this paper involves all stages of system development, including requirement analysis, data model design, coding implementation, etc., and the feasibility and effectiveness of the design scheme is verified through system testing. The realization of the system provides a convenient, efficient and real-time feedback channel for colleges and universities, and provides reference for the development and improvement of college network suggestion box.