摘要
政务信息可反映单位整体的工作情况,是检验单位工作扎实程度、服务水平高低的一个指标,抓好政务信息质量对提升政务工作有着重要的意义。政务信息中的风险内容具有数量大、与日常表述习惯不一致的特点,信息员在编写信息时难以自主规避此类问题。本文以山西省地震局政务网站近五年的政务信息为研究对象,分析风险内容的类型和特点,构建本系统的风险内容词库,同时结合自然语言处理和智能云计算平台,实现政务信息风险内容的自动识别,为官网信息的规范化、提高政务网站内容的质量提供探索思路与方法。
Government information reflects the overall work situation of an organization and serves as an indicator of the solidity of its operations and the level of its services.Ensuring the quality of government information is significant for improving governmental work.The risk content in government information is characterized by its large volume and inconsistency with usual expression habits,making it difficult for information officers to autonomously avoid such issues when compiling information.This paper focuses on the governmental information from the Shanxi Earthquake Agency website over the past five years,analyzing the types and features of risk-related content,and establishes a risk content lexicon for this system.Additionally,by integrating natural language processing and intelligent cloud computing platforms,it achieves automatic identification of risk content in governmental information.This provides exploratory thoughts and methods for standardizing information on the official website and enhancing the quality of content on government websites.
作者
王宁
闫晓美
张瑞芳
许振鹏
WANG Ning;YAN Xiaomei;ZHANG Ruifang;XU Zhenpeng(Shanxi Earthquake Agency,Taiyuan,Shanxi 030021,China;National Continental Rift Valley Dynamics Observatory of Taiyuan,Taiyuan,Shanxi 030025,China)
出处
《山西地震》
2024年第2期48-52,共5页
Earthquake Research in Shanxi
基金
山西省地震局科研项目(SBK—2313)。
关键词
政务网站
政务信息
风险
自动识别
government website
government information
risk
automatic
