School of Traffic and Transportation Engineering, Changsha University of Science and Technology, Changsha 410114, China.
Key Laboratory of Highway Engineering (Changsha University of Science & Technology), Ministry of Education, Changsha, China.
Comput Intell Neurosci. 2022 Jul 13;2022:2897672. doi: 10.1155/2022/2897672. eCollection 2022.
To effectively diagnose and monitor the vertical collusion in construction project bidding, this paper developed a comprehensive evaluation model with deep neural network and transfer learning. By this model, the collusion characteristics of bidders, tenderers, and bid evaluation experts were mined from limited data set hidden and collusion tendency was evaluated. Firstly, 18 evaluation indicators were established from literature review, court file summarization, typical case analysis, and expert consultation. Then, a comprehensive evaluation model was developed with the deep neural network and transfer learning. Finally, the model was trained and tested with the collected data set. The test results showed that the developed model achieved 87.3% identification accuracy in collusion tendency evaluation of different subjects.
为了有效诊断和监测建设工程招标投标中的串通行为，本文开发了一个基于深度学习神经网络和迁移学习的综合评价模型。通过该模型，从有限的数据集隐藏信息中挖掘投标人、招标人、评标专家的串通特征，并评价串通倾向。首先，从文献回顾、法院档案总结、典型案例分析和专家咨询中确定了 18 个评价指标。然后，利用深度神经网络和迁移学习建立了综合评价模型。最后，利用收集到的数据集对模型进行了训练和测试。测试结果表明，该模型在不同主体的串通倾向评价中达到了 87.3%的识别准确率。
