{"id":98345,"updated":"2025-01-21T12:29:44.784242+00:00","links":{},"created":"2025-01-18T23:44:37.089793+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00098345","sets":["581:7397:7440"]},"path":["7440"],"owner":"11","recid":"98345","title":["eテスティングにおけるLDAを用いた項目間類似度の算出"],"pubdate":{"attribute_name":"公開日","attribute_value":"2014-01-15"},"_buckets":{"deposit":"fd6c17d4-a802-4ae0-9168-01eee5d8140f"},"_deposit":{"id":"98345","pid":{"type":"depid","value":"98345","revision_id":0},"owners":[11],"status":"published","created_by":11},"item_title":"eテスティングにおけるLDAを用いた項目間類似度の算出","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"eテスティングにおけるLDAを用いた項目間類似度の算出"},{"subitem_title":"Calculating Test Item Similarity Using Latent Dirichlet Allocation in E-Testing","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"[特集：教育とコンピュータ] eテスティング，アイテム・バンク，類似項目，LDA（Latent Dirichlet Allocation），類似度","subitem_subject_scheme":"Other"}]},"item_type_id":"2","publish_date":"2014-01-15","item_2_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"電気通信大学大学院情報システム学研究科／日本学術振興会特別研究員 PD"},{"subitem_text_value":"岩手県立大学ソフトウェア情報学部"},{"subitem_text_value":"東京電機大学未来科学部"},{"subitem_text_value":"電気通信大学大学院情報システム学研究科"}]},"item_2_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Graduate School of Information Systems, The University of Electro-Communications / Research Fellow of Japan Society for the Promotion of Science","subitem_text_language":"en"},{"subitem_text_value":"Faculty of Software and Information Science, Iwate Prefectural University","subitem_text_language":"en"},{"subitem_text_value":"School of Science and Technology for Future Life, Tokyo Denki University","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Information Systems, The University of Electro-Communications","subitem_text_language":"en"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"publish_status":"0","weko_shared_id":-1,"item_file_price":{"attribute_name":"Billing file","attribute_type":"file","attribute_value_mlt":[{"url":{"url":"https://ipsj.ixsq.nii.ac.jp/record/98345/files/IPSJ-JNL5501011.pdf"},"date":[{"dateType":"Available","dateValue":"2016-01-15"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-JNL5501011.pdf","filesize":[{"value":"1.1 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"660","billingrole":"5"},{"tax":["include_tax"],"price":"330","billingrole":"6"},{"tax":["include_tax"],"price":"0","billingrole":"8"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"f2123311-420c-4e51-8a6d-4609d918c46a","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2014 by the Information Processing Society of Japan"}]},"item_2_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"高木, 輝彦"},{"creatorName":"高木, 正則"},{"creatorName":"勅使河原, 可海"},{"creatorName":"田中, 健次"}],"nameIdentifiers":[{}]}]},"item_2_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Teruhiko, Takagi","creatorNameLang":"en"},{"creatorName":"Masanori, Takagi","creatorNameLang":"en"},{"creatorName":"Yoshimi, Teshigawara","creatorNameLang":"en"},{"creatorName":"Kenji, Tanaka","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_2_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN00116647","subitem_source_identifier_type":"NCID"}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_6501","resourcetype":"journal article"}]},"item_2_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"1882-7764","subitem_source_identifier_type":"ISSN"}]},"item_2_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"本研究ではこれまで，eテスティングにおいて類似項目を自動検索することを目的とし，項目間類似度の算出手法を提案，実験・評価を行ってきた．類似度データや類似項目を用いることで，(1)類似項目の自動検索，(2)自動的なアイテム・バンクの構築，(3)項目間構造の可視化，(4)テスト情報量の向上，(5)新規項目の難易度の推定，(6)適応的なテストの出題，(7)項目の作成，などの支援が可能となる．本論文では，項目間類似度の算出精度の向上を目的とし，文書の生成過程を確率的にモデル化したLDAを適用した新たな手法を提案する．提案手法では，項目ごとにLDAで推定されたトピックを基に特徴ベクトルを生成し，余弦により項目間類似度を算出する．LDAを適用することにより，(1)不要な単語による誤検索の解消や，(2)項目の内容理解に踏み込んだ特徴ベクトルの生成，が期待される．提案手法のオリジナルなアイディアは，項目ごとにLDAでトピックを推定する際に，(a)重要な語が出現する箇所を自動で決定し，(b)限られた語の共起性を高める，という前処理にある．初級システムアドミニストレータ試験で出題された250項目を対象とした，類似項目の検索実験を行った結果，提案手法ではLDAを用いたその他の比較手法や既存手法に比べ，項目間類似度の算出精度が最も高かった．これらの実験・評価結果から，前処理(a)，(b)の有効性や項目間類似度の算出にLDAを適用することの有効性が示唆された．","subitem_description_type":"Other"}]},"item_2_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"In previous studies, to retrieve similar item automatically in e-testing, we proposed methods of calculating similarity between items, and conducted experiments and evaluations. It is possible to apply similarity data or similar item to (1) automatically retrieving similar items, (2) automatically constructing item banks, (3) visualizing structure between items, (4) optimizing the amount of test information, (5) estimating the difficulty level of unanswered items, (6) computer adaptive testing (CAT) and (7) supporting the creation of items. In this paper, to improve the accuracy of calculating similarity between items, we propose new method of calculating similarity between items applied Latent Dirichlet Allocation (LDA), a generative probabilistic document model. Fundamentally, we assume that each item is represented by a vector using topics estimated by LDA, and the similarity between items is calculated by cosine. Applying LDA to calculating similarity between items provides (1) decreasing the number of retrieved dissimilar items, and (2) creating vectors based of the relation between extracted terms. To accurately estimate topics in each item, we take the two following approaches: (a) Identifying the part in which the important term occurs. (b) Enhancing the co-occurrence relation between terms. We targeted 250 items tested by Systems Administrator Examination and conducted experiment which similar items are retrieved. The result of experiment showed the effectiveness of approach (a), (b) and applying LDA to calculating similarity between items. We furthermore demonstrated the improvement in accuracy of the proposed method in comparison with existing methods.","subitem_description_type":"Other"}]},"item_2_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"104","bibliographic_titles":[{"bibliographic_title":"情報処理学会論文誌"}],"bibliographicPageStart":"91","bibliographicIssueDates":{"bibliographicIssueDate":"2014-01-15","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"1","bibliographicVolumeNumber":"55"}]},"relation_version_is_last":true,"weko_creator_id":"11"}}