基于爬虫和深度学习的计量信息数据推送系统

    Metrology Information Data Push System Based on Crawler and Deep Learning

    • 摘要: 国家计量科学数据中心开发了计量信息数据推送系统。该系统利用爬虫技术从国内外计量权威网站抓取最新计量信息数据,经AI预处理及人工审核后构建到三个维度的主题地图,再发布到网站,推送给相应主题词订阅用户。科研人员可按三个维度的主题地图或自定义主题词进行信息订阅,实时获取最新领域的计量信息数据。该系统已在国家计量科学数据中心网站上线,实际运行效果较好,验证了方案的可行性和发展潜力。

       

      Abstract: The National Metrology Data Center has developed a metrology information data push system. The system first uses the web crawler technology to capture the latest metrology information data from authoritative metrology websites at home and abroad. Then the data is integrated to a three-dimensional topic map after being preprocessed by AI and manually curated, which is further published to the website and pushed to the corresponding subscribers with corresponding subject terms. Researchers can subscribe to the information according to the three dimensional topic maps or custom subject terms to obtain the latest metrology information data in real time. The system has been deployed on the website of the National Metrology Data Center, and the actual operation is good, which verifies the feasibility and development potential of the program.

       

    /

    返回文章
    返回