2026-03-12T01:51:24Z https://ipsj.ixsq.nii.ac.jp/oai

oai:ipsj.ixsq.nii.ac.jp:00211799 2025-01-19T17:40:19Z 1164:3782:10507:10595

Generating RDF Metadata from Twitter Streams Generating RDF Metadata from Twitter Streams Teklu, Aregawi Gidey Toshiyuki, Amagasa Teklu, Aregawi Gidey Toshiyuki, Amagasa DC In recent years, Social Network Service has gained much public attention as a source of information. Twitter is considered as most prominent SNS source. Most of previous studies on the content of tweets have focused on the detection of emerging topics and some have focused on creation of RDF. It is important to generate RDF metadata from tweets to put data in context via linking and semantic metadata providing a framework for data integration, analytics and sharing. However, in the previous studies, the generated RDFs were not from streaming tweets and the mentions interlinked with knowledge graphs did not show the location of the text in the tweet. In our experiment, we generated RDF metadata from streaming tweets using a hashtag 'Covid-19' and we located the exact position of the mentions in the tweet being linked to a knowledge graph, WikiPedia using TagMe entity linker. We expressed it as a URI which includes the URL of Twitter, the user name, the tweet id and index of the offset information. In recent years, Social Network Service has gained much public attention as a source of information. Twitter is considered as most prominent SNS source. Most of previous studies on the content of tweets have focused on the detection of emerging topics and some have focused on creation of RDF. It is important to generate RDF metadata from tweets to put data in context via linking and semantic metadata providing a framework for data integration, analytics and sharing. However, in the previous studies, the generated RDFs were not from streaming tweets and the mentions interlinked with knowledge graphs did not show the location of the text in the tweet. In our experiment, we generated RDF metadata from streaming tweets using a hashtag 'Covid-19' and we located the exact position of the mentions in the tweet being linked to a knowledge graph, WikiPedia using TagMe entity linker. We expressed it as a URI which includes the URL of Twitter, the user name, the tweet id and index of the offset information. 情報処理学会 2021-07-08 eng technical report https://ipsj.ixsq.nii.ac.jp/records/211799 2188-8892 AN10539261 研究報告ドキュメントコミュニケーション（DC） 2021-DC-121 7 1 8 https://ipsj.ixsq.nii.ac.jp/record/211799/files/IPSJ-DC21121007.pdf application/pdf 2.7 MB 2023-07-08