计算机技术与发展2026，Vol.36Issue(1)：184-192,9.DOI:10.20165/j.cnki.ISSN1673-629X.2025.0192
基于深度学习的高速公路数据质量管理研究
Research on Highway Data Quality Management Based on Deep Learning
摘要
Abstract
With the rapid development of informatization on expressways,the effective management and strategic utilization of operational data have emerged as pivotal factors in augmenting both the service quality and operational efficacy of the road network infrastructure.We construct a systematic highway operation data quality management methodology framework which consists of three core modules:data quality assessment,data quality analysis,and data quality repair.Firstly,a multi-dimensional operational data quality evaluation model is constructed using the Analytic Hierarchy Process(AHP)combined with the CRITIC,enabling quantitative assessment of evaluation indi-cators.Secondly,a data quality analysis model based on the isolation forest algorithm is established to effectively identify and locate data anomalies.Finally,a novel data quality repair model is constructed based on the Temporal Convolutional Network(TCN)optimized by the Self-Attention(SA)mechanism,which utilizes deep learning techniques to realize the intelligent repair of data outliers.To validate the repair effectiveness,we input both pre-repair and post-repair data into multiple regression prediction models for comparative analysis.The results demonstrate significant improvements in the predictive performance of all models.The implementation of this framework not only effectively enhances the quality of expressway operational data and provides a reliable data foundation for operational decision support systems,but also holds substantial practical significance for promoting the digital transformation of the expressway industry.关键词
交通数据/数据质量管理/孤立森林算法/时间卷积网络/自注意力机制Key words
traffic data/data quality management/isolated forest algorithm/temporal convolutional network/self-attention mechanism分类
信息技术与安全科学引用本文复制引用
孙楠,陈思行,刘烨,黄毅嵩..基于深度学习的高速公路数据质量管理研究[J].计算机技术与发展,2026,36(1):184-192,9.基金项目
陕西省创新能力支撑计划项目(2024RS-CXTD-27) （2024RS-CXTD-27）
