Design of a University Book Push System Based on Big Data

In response to the existing university book recommendation systems that mostly recommend readers based on library resources and lack exploration of recommended books, this paper proposes a system service architecture that accurately pushes both readers and libraries simultaneously. The system is designed using technologies such as web spiders, big data, NLP, and visual analysis. The system integrates reader data entry, web crawler, user profile, book similarity measurement, and intelligent push functions. Book managers can use the system to push books accurately and discover recommended books. The system fully utilizes book attribute data and reader attribute data, avoiding common cold start issues in recommendation systems, and providing support for improving the service level of the library and scientifically allocating collection resources.