检索规则说明:AND代表“并且”;OR代表“或者”;NOT代表“不包含”;(注意必须大写,运算符两边需空一格)
检 索 范 例 :范例一: (K=图书馆学 OR K=情报学) AND A=范并思 范例二:J=计算机应用与软件 AND (U=C++ OR U=Basic) NOT M=Visual
作 者:王菁[1,2] 张睿轩 邓欣颖 林浩然 WANG Jing;ZHANG Ruixuan;DENG Xinying;LIN Haoran(School of Information Science and Technology North China University of Technology,Beijing 100144,China;Beijing Key Laboratory on Integration and Analysis of Large-Scale Stream Data,Beijing 100144,China)
机构地区:[1]北方工业大学信息学院,北京100144 [2]大规模流数据集成与分析技术北京市重点实验室,北京100144
出 处:《北方工业大学学报》2024年第5期30-38,共9页Journal of North China University of Technology
基 金:国家自然科学基金国际(地区)合作与交流项目(62061136006)。
摘 要:地址是我国城市数字化建设的重要空间资源和战略性基础信息资源,经解析处理后可应用在多个领域,因此提升地址解析的准确性十分必要。针对以上问题,本文提出一种基于深度学习模型的中文地址解析方法。该方法首先使用预训练语言模型来获取动态词向量,并利用双向门控循环单元的序列建模能力和卷积神经网络的平移不变性提取文本中的上下文语义特征;随后引入层归一化规范模型训练过程中的特征分布,再添加多头注意力机制使模型能够更有效地聚焦关键信息;最后引入条件随机场用于全局标注序列的学习和解码,完成中文地址解析任务。实验结果表明,所提模型在中文地址解析任务中的精确率、召回率和F1值均高于其他对比中文地址解析模型。Addresses are crucial spatial resources and strategic foundational information in the digitalization of cities in our country.After being parsed and processed,they can be applied in various fields,making the improvement of address parsing accuracy highly necessary.To address this issue,this paper proposes a method for Chinese address parsing based on a deep learning model.The proposed method initially employs a pre-trained language model to obtain dynamic word embeddings.It then utilizes the sequence modeling capability of Bidirectional Gated Recurrent Units(BiGRU)and the translational invariance of Convolutional Neural Networks(CNN)to extract contextual semantic features from the text.Subsequently,layer normalization is introduced to regulate feature distribution during model training.The model further incorporates a multi-head attention mechanism to enable more effective focus on key information.Finally,Conditional Random Fields(CNN)to applied for global sequence labeling learning and decoding to complete the Chinese address parsing task.Experimental results demonstrate that the proposed model achieves higher precision,recall,and Fl score in Chinese address parsing tasks compared to other existing models.
关 键 词:地址解析 地址要素分类 地址解析模型 预训练语言模型 双向门控循环单元
分 类 号:TP393.1[自动化与计算机技术—计算机应用技术]
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在链接到云南高校图书馆文献保障联盟下载...
云南高校图书馆联盟文献共享服务平台 版权所有©
您的IP:3.148.192.32