深度森林融合模型优化高校电子图书采购策略

Optimizing University Electronic Book Procurement Strategy with Deep Forest Fusion Model

摘要: 随着高校对电子图书采购需求的明显增加，为提升采购决策效能，文章提出了一种深度森林融合算法，即LightGBM和CatBoost融合为LHGCAT-XDF的优化模型。该模型兼具LightGBM低内存消耗、和CatBoost低时间复杂度的特点。通过实验结果显示，LHGCAT-XDF相较传统机器学习模型在综合性能上更为卓越，有效克服了传统采购模型在精准性和效率方面的限制，为高校图书馆电子图书采购提供可靠的决策支持。

Abstract: With the evident increase in demand for electronic book procurement in universities, this paper proposes an optimized model, LHGCAT-XDF, by integrating the LightGBM and CatBoost algorithms to enhance the efficiency of procurement decision-making. This model has the characteristics of low memory consumption of LightGBM and low time complexity of CatBoost. Experimental results demonstrate that LHGCAT-XDF outperforms traditional machine learning models in comprehensive performance, effectively overcoming the limitations of traditional procurement models in precision and efficiency. This provides reliable decision support for electronic book procurement in university libraries.