文献类型：期刊文献

英文题名：Intelligent Retrieval System for Ship Fault Information Based on Big Data Analysis

作者：Sheng, Hongyu[1];Wang, Xinqiang[2]

第一作者：盛鸿宇

通讯作者：Wang, XQ[1]

机构：[1]Beijing Union Univ, Coll Robot, Beijing 100101, Peoples R China;[2]Tianjin Sino German Univ Appl Sci, Sch Software & Commun, Tianjin 300350, Peoples R China

第一机构：北京联合大学机器人学院

通讯机构：[1]corresponding author), Tianjin Sino German Univ Appl Sci, Sch Software & Commun, Tianjin 300350, Peoples R China.

年份：2019

卷号：93

期号：sp1

起止页码：1019-1025

外文期刊名：JOURNAL OF COASTAL RESEARCH

收录：;Scopus(收录号：2-s2.0-85072657617);WOS:【SCI-EXPANDED(收录号:WOS:000487997100148)】；

基金：This work was supported by National Project of "Innovation and Practice of Cooperative Educational Model of New Subjects, Schools and Enterprises in Open, Fused and Shared Local Colleges and Universities", Higher Education Office Letter No. 17 (2018); Tianjin Municipal Natural Science Foundation (17JCQNJC00500); Tianjin Municipal Science and Technology Commissioner Project (17JCTPJC50600, 18JCTPJC61800).

语种：英文

外文关键词：Big data analysis; ship fault; intelligent retrieval; Lucene retrieval component

摘要：In order to improve the efficiency of ship fault information retrieval and shorten the retrieval time, an intelligent retrieval system for ship fault information based on big data analysis is designed. The Lucene retrieval component is used to build a multi-level classification module, which matches the query keywords input in the process of ship fault information retrieval with the ship operation status to increase the intelligent retrieval conditions. Combining the big data analysis technology to classify the ship fault information at different levels, the tree classification structure is presented in the process of ship fault status classification, which can refine the classification of ship fault, describe the attributes of retrieval objects in detail, reduce the number of retrieval times, and improve the efficiency of ship fault status retrieval, thus completing the design of the intelligent retrieval system for ship fault information. The experimental results show that under the same conditions, the retrieval time of the system is nearly half shorter than that of the traditional system, and the longest retrieval time can be controlled within 2 seconds, which effectively improves the retrieval efficiency.