为了自动、高效地数字化处理船舶设计、制造和维修过程中所生成的海量资料——图纸和文件,对数字化处理过程中的关键技术——著录进行了研究,提出了快速批量著录的概念,并采用数据库技术构建了快速批量著录系统。针对快速批量著录系统中的瓶颈技术——文本自动标引,结合船舶资料的特点和规律,提出并实现了基于统计原理的位置权重方法,有效地提高了文本自动标引的效率和准确度。在此基础上,研制出数字化处理平台,实现船舶资料扫描、识别、著录、输出、共享和管理等功能。
For solving the digitalization problems of huge drawings and files in ships, the key technology of fast and batch record for these drawings and files have been researched, and the concept and method of fast and batch record has been made definite. Because the technology of text auto index is the most difficult, so the position weight scheme methods have been realized within characteristics and laws of ship drawing and file, and also the efficiency and accuracy of text auto index have been effectively raised. Finally, on the basis of above-mentioned works, the digitalization system for ship drawing and file has been developed, this system has many functions such as scanning, distinguish, record, output, shared and manage, etc.
2019,41(7): 134-136 收稿日期:2018-07-04
DOI:10.3404/j.issn.1672-7649.2019.07.026
分类号:G254
作者简介:马曲立(1962-),男,教授,主要研究方向为装备保障指挥
参考文献:
[1] 刘艳文, 周朝晖. 自动标引中船舶资料位置权重方案的确定[J]. 科技情报开发与经济, 2012, 22(17):101-104
[2] 李千驹, 李思达, 刘建毅. 一种基于知识组织的关键词自动标引方法[J]. 情报科学, 2016, 34(11):107-110
[3] FABRIZIO Sebastiani. Machine learning in automated text categorization[J]. ACM Computing Surveys, 2002, 34(1):11-33
[4] 刘艳文, 周朝晖, 向晖. 船舶图纸资料数字化平台关键技术的实现[J]. 机械工程师, 2012, 11:1-3
[5] 万莉. 学术期刊知识交流效率评价及影响因素研究[J]. 中国科技期刊研究, 2017, 28(12):1160-1165
[6] 孙东莹. 电子图书和纸质图书的著录异同[J]. 江苏科技信息, 2016, 1(3):9-10
[7] 马向东. 一种形式与内容相结合的多媒体分类方法研究与实现[J]. 经济研究导刊, 2016, 23(304):142-143
[8] 蒋宏毅, 王红蕾, 边鹏飞, 等. 地震模拟图纸数字化存储的实现[J]. 地震地磁观测与研究, 2015, 6(3):133-138