交通信息与安全2025，Vol.43Issue(3)：33-43,11.DOI:10.3963/j.jssn.1674-4861.2025.03.004
基于多模态信息融合的铁路安全知识深度挖掘与生成式推荐方法
Deep Mining and Association Recommendation Method for Railway Safety Knowledge Based on Multimodal Information Fusion
摘要
Abstract
The rapid digital and intelligent transformation of railway information systems has created an urgent de-mand for fine-grained,explainable safety knowledge recommendations.To address the fragmentation of cross-mod-al associations and insufficient alignment with operational rules exhibited by traditional approaches,a framework in-tegrating multimodal feature fusion with generative reasoning is investigated.A hierarchical railway safety knowl-edge graph is constructed,and topological features under business constraints are extracted via the Node2Vec algo-rithm.Simultaneously,a lightweight Transformer encoder(GTE)captured deep semantic embeddings of individual safety clauses.To balance contributions from graph and text features,a tunable weighting strategy is introduced,dy-namically controlling the fusion ratio of text vectors and graph embeddings and applying a dual-constraint mecha-nism based on cosine similarity and predefined rules to generate candidate recommendations.A three-stage progres-sive retrieval architecture is designed to achieve precise multimodal alignment and suppress noise.Finally,the Deep-Seek-R1 large language model served as the reasoning engine,with domain-specific prompting converting retrieved candidates into executable decision plans,thereby enhancing coherence and interpretability.Experiments on 27 safe-ty documents from a railway operator,using a similarity threshold of 0.85 and a maximum of 10 recommendations per query,demonstrated a recommendation accuracy of 95%(an 8-percentage-point improvement over traditional methods)along with significant gains in contextual relevance and explainability.This investigation confirms the syn-ergistic benefits of multimodal retrieval and generative reasoning,providing a robust technical foundation for evolv-ing railway safety knowledge services from precise recommendation to intelligent decision support.关键词
铁路安全知识推荐框架/多模态特征融合/知识图谱/生成式推理/文本关联分析Key words
railway safety knowledge recommendation framework/multimodal feature fusion/knowledge graph/generative reasoning/textual association analysis分类
交通工程引用本文复制引用
高丽,杨诺晗,李晴,王永恒,严晗,赵汝豪,马小平..基于多模态信息融合的铁路安全知识深度挖掘与生成式推荐方法[J].交通信息与安全,2025,43(3):33-43,11.基金项目
国家自然科学基金青年项目(61903023)、社会科学横向项目(B24SK00250)资助 （61903023）
