With the rapid growth of technologies such as artificial intelligence (AI), big data, and cloud computing, education informatization and modernization have received strong technical support, and the education industry is gradually entering a new stage of smart education. Education big data, as an important collection of data reflecting the current situation, needs, and problems of basic education resource (ER) allocation, has become a key tool for reshaping the structure of ER allocation and building a high-quality education ecosystem. However, the problem of uneven distribution of ER and the increasing demand for personalized learning remains prominent. In response to this situation, this article proposes an educational resource optimization configuration and management model based on particle swarm optimization algorithm (PSO) and deep learning (DL), combined with big data analysis technology. This model aims to achieve rational allocation of ER and provide personalized educational resource recommendations to students through DL technology, thereby maximizing resource utilization efficiency. The results shows that the model not only significantly improves the utilization rate of ER, but also performs well in the accuracy of personalized recommendations.