Co-occurrence technique and dictionary based method for Indonesian thesaurus construction

2017 5th International Conference on Information and Communication Technology (ICoIC7) Pub Date : 2017-05-01 DOI:10.1109/ICOICT.2017.8074649

R. W. Sholikah, A. Arifin, D. Purwitasari, C. Fatichah

{"title":"Co-occurrence technique and dictionary based method for Indonesian thesaurus construction","authors":"R. W. Sholikah, A. Arifin, D. Purwitasari, C. Fatichah","doi":"10.1109/ICOICT.2017.8074649","DOIUrl":null,"url":null,"abstract":"Thesaurus as control vocabulary can be an important tool in Natural Language Processing (NLP). However, constructing a thesaurus manually by experts can be time consuming. Besides that the subjectivity of each expert can affect the structure of the thesaurus. A lot of method has already been implemented to build an automatic thesaurus in languages that categorized as rich language resources. In poor language resources such as Indonesia, the research about this field is still limited. This paper proposed a framework to construct a thesaurus in Indonesian language using monolingual corpus. The method will use Indonesian dictionary and large monolingual corpus from news articles. The candidate related terms will be extracted from every resource, then the two candidate will produce the final result of thesaurus. The evaluation is done by using the thesaurus as QE (Query Expansion) resource in IR (Information Retrieval) system. The experimental results show that using the automatic thesaurus can obtain the precision and recall of the system with 54.00% and 85.42%, respectively.","PeriodicalId":244500,"journal":{"name":"2017 5th International Conference on Information and Communication Technology (ICoIC7)","volume":"78 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 5th International Conference on Information and Communication Technology (ICoIC7)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICOICT.2017.8074649","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 1

Abstract

Thesaurus as control vocabulary can be an important tool in Natural Language Processing (NLP). However, constructing a thesaurus manually by experts can be time consuming. Besides that the subjectivity of each expert can affect the structure of the thesaurus. A lot of method has already been implemented to build an automatic thesaurus in languages that categorized as rich language resources. In poor language resources such as Indonesia, the research about this field is still limited. This paper proposed a framework to construct a thesaurus in Indonesian language using monolingual corpus. The method will use Indonesian dictionary and large monolingual corpus from news articles. The candidate related terms will be extracted from every resource, then the two candidate will produce the final result of thesaurus. The evaluation is done by using the thesaurus as QE (Query Expansion) resource in IR (Information Retrieval) system. The experimental results show that using the automatic thesaurus can obtain the precision and recall of the system with 54.00% and 85.42%, respectively.

查看原文本刊更多论文

印尼语同义词典构建的共现技术和基于词典的方法

同义词典作为控制词汇是自然语言处理(NLP)的重要工具。然而，由专家手动构建同义词库可能很耗时。此外，每个专家的主观性会影响同义词典的结构。目前已经实现了许多方法，可以在被归类为丰富语言资源的语言中构建自动同义词库。在印度尼西亚等语言资源贫乏的国家，这一领域的研究仍然有限。本文提出了一个利用单语语料库构建印尼语词库的框架。该方法将使用印尼语词典和新闻文章中的大型单语语料库。从每个资源中提取候选相关术语，然后两个候选将产生同义词库的最终结果。在信息检索系统中，利用词库作为查询扩展资源进行评价。实验结果表明，使用自动词库，系统的查准率和查全率分别达到54.00%和85.42%。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

2017 5th International Conference on Information and Communication Technology (ICoIC7)

自引率

0.00%

发文量