Under the premise of coordinated procurement bilateral and multi-issue negotiation, adaptive negotiation strategy has become an essential factor for multiagent conflict resolution. This paper studies an adaptive negotiation strategy based on selective integrated learning, which effectively improves negotiation. First, take the suppliers and purchasing companies in the cluster supply chain as the research objects and analyze the characteristics of multilateral negotiation of collaborative procurement. Secondly, the support vector machine algorithm performs adaptive learning for each evaluation data set to estimate the concession range. On this basis, remove the few submodels that perform poorly, recombine the calculation weights, and establish a multiagent clustered supply collaborative procurement negotiation model. The simulation experiment proves the feasibility of the adaptive negotiation strategy and the effectiveness of the adaptive coordination strategy based on selective ensemble learning proposed in this paper from the aspects of concession range prediction error rate, prediction accuracy rate, and negotiation utility.

中文翻译：

---

自创区产业集群集成学习研究

在协调采购双边和多问题谈判的前提下，自适应谈判策略已成为多主体冲突解决的重要因素。本文研究了一种基于选择性集成学习的自适应谈判策略，有效提高了谈判效率。首先，以集群供应链中的供应商和采购公司为研究对象，分析协同采购多边谈判的特点。其次，支持向量机算法对每个评价数据集进行自适应学习，估计让步范围。在此基础上，剔除少数表现不佳的子模型，重新组合计算权重，建立多智能体集群供应协同采购谈判模型。