{"_buckets": {"deposit": "73839b68-c826-41a7-ae81-27c4d69ad51e"}, "_deposit": {"created_by": 15, "id": "8383", "owners": [15], "pid": {"revision_id": 0, "type": "depid", "value": "8383"}, "status": "published"}, "_oai": {"id": "oai:mie-u.repo.nii.ac.jp:00008383", "sets": ["639"]}, "author_link": [], "item_10001_biblio_info_7": {"attribute_name": "書誌情報", "attribute_value_mlt": [{"bibliographicIssueDates": {"bibliographicIssueDate": "1999-04-25", "bibliographicIssueDateType": "Issued"}, "bibliographicIssueNumber": "4", "bibliographicPageEnd": "630", "bibliographicPageStart": "621", "bibliographicVolumeNumber": "J82-D-II", "bibliographic_titles": [{"bibliographic_title": "電子情報通信学会論文誌. D-II, 情報・システム, II-パターン処理 = The transactions of the Institute of Electronics, Information and Communication Engineers. D-II"}]}]}, "item_10001_description_19": {"attribute_name": "フォーマット", "attribute_value_mlt": [{"subitem_description": "application/pdf", "subitem_description_type": "Other"}]}, "item_10001_description_5": {"attribute_name": "抄録", "attribute_value_mlt": [{"subitem_description": "学習標本の大きさが有限の場合, 2次識別関数の識別性能は特徴ベクトルの次元数の増加に伴って劣化しピーキング現象を起こすことが知られている. このような問題に対処するために, ベイズアプローチによって導かれる未知の分布に対する最適識別関数の平均誤り確率等を理論的・実験的に評価して, 学習標本の大きさがクラス間で等しい場合と異なる場合の最適識別系の有効性を明らかにする. その結果, 学習標本の大きさがほぼ等しい場合は, 最適識別関数及び最適識別関数から導かれる擬似ベイズ識別関数, 改良投影距離, 投影距離を用いる識別系ではピーキング現象が起きないことがわかった. また, 学習標本の大きさが異なる場合は, 最適識別関数及び擬似ベイズ識別関数ではピーキング現象が起きないが, 2次識別関数, 改良投影距離, 投影距離を用いる識別系ではピーキング現象が起きることがわかった.", "subitem_description_type": "Abstract"}]}, "item_10001_publisher_8": {"attribute_name": "出版者", "attribute_value_mlt": [{"subitem_publisher": "電子情報通信学会"}]}, "item_10001_relation_42": {"attribute_name": "関係URI", "attribute_value_mlt": [{"subitem_relation_name": [{"subitem_relation_name_text": "http://search.ieice.org/"}]}]}, "item_10001_source_id_11": {"attribute_name": "書誌レコードID", "attribute_value_mlt": [{"subitem_source_identifier": "AA11340957", "subitem_source_identifier_type": "NCID"}]}, "item_10001_source_id_9": {"attribute_name": "ISSN", "attribute_value_mlt": [{"subitem_source_identifier": "0915-1923", "subitem_source_identifier_type": "PISSN"}]}, "item_10001_subject_21": {"attribute_name": "日本十進分類法", "attribute_value_mlt": [{"subitem_subject": "007", "subitem_subject_scheme": "NDC"}]}, "item_10001_text_25": {"attribute_name": "その他のタイトル", "attribute_value_mlt": [{"subitem_text_value": "On the limited sample effect of the optimum classifier by bayesian approach : the case of independent sample size for each class"}]}, "item_10001_text_68": {"attribute_name": "ノート", "attribute_value_mlt": [{"subitem_text_value": "copyright©1999IEICE"}]}, "item_10001_text_70": {"attribute_name": "資源タイプ（三重大）", "attribute_value_mlt": [{"subitem_text_value": "Journal Article / 学術雑誌論文"}]}, "item_10001_version_type_20": {"attribute_name": "著者版フラグ", "attribute_value_mlt": [{"subitem_version_resource": "http://purl.org/coar/version/c_970fb48d4fbd8a85", "subitem_version_type": "VoR"}]}, "item_creator": {"attribute_name": "著者", "attribute_type": "creator", "attribute_value_mlt": [{"creatorNames": [{"creatorName": "韓, 雪仙", "creatorNameLang": "ja"}, {"creatorName": "Han, Xuexian", "creatorNameLang": "en"}]}, {"creatorNames": [{"creatorName": "若林, 哲史", "creatorNameLang": "ja"}, {"creatorName": "Wakabayashi, Tetsushi", "creatorNameLang": "en"}]}, {"creatorNames": [{"creatorName": "木村, 文隆", "creatorNameLang": "ja"}], "familyNames": [{"familyName": "Kimura, Fumitaka", "familyNameLang": "en"}]}, {"creatorNames": [{"creatorName": "三宅, 康二", "creatorNameLang": "ja"}, {"creatorName": "Miyake, Yasuji", "creatorNameLang": "en"}]}]}, "item_files": {"attribute_name": "ファイル情報", "attribute_type": "file", "attribute_value_mlt": [{"accessrole": "open_date", "date": [{"dateType": "Available", "dateValue": "2017-02-18"}], "displaytype": "detail", "download_preview_message": "", "file_order": 0, "filename": "40A12251.pdf", "filesize": [{"value": "307.8 kB"}], "format": "application/pdf", "future_date_message": "", "is_thumbnail": false, "licensetype": "license_note", "mimetype": "application/pdf", "size": 307800.0, "url": {"label": "40A12251.pdf", "url": "https://mie-u.repo.nii.ac.jp/record/8383/files/40A12251.pdf"}, "version_id": "83721094-ca8f-46d7-b167-d35f0f7e22b2"}]}, "item_keyword": {"attribute_name": "キーワード", "attribute_value_mlt": [{"subitem_subject": "ベイズアプローチ", "subitem_subject_scheme": "Other"}, {"subitem_subject": "ベイズ学習", "subitem_subject_scheme": "Other"}, {"subitem_subject": "最適識別関数", "subitem_subject_scheme": "Other"}, {"subitem_subject": "投影距離", "subitem_subject_scheme": "Other"}, {"subitem_subject": "平均誤り確率", "subitem_subject_scheme": "Other"}, {"subitem_subject": "手書き数字認識", "subitem_subject_scheme": "Other"}]}, "item_language": {"attribute_name": "言語", "attribute_value_mlt": [{"subitem_language": "jpn"}]}, "item_resource_type": {"attribute_name": "資源タイプ", "attribute_value_mlt": [{"resourcetype": "journal article", "resourceuri": "http://purl.org/coar/resource_type/c_6501"}]}, "item_title": "ベイズアプローチによる最適識別系の有限標本効果に関する考察 : 学習標本の大きさがクラス間で異なる場合(\u003c論文小特集\u003eパターン認識のための学習 : 基礎と応用)", "item_titles": {"attribute_name": "タイトル", "attribute_value_mlt": [{"subitem_title": "ベイズアプローチによる最適識別系の有限標本効果に関する考察 : 学習標本の大きさがクラス間で異なる場合(\u003c論文小特集\u003eパターン認識のための学習 : 基礎と応用)", "subitem_title_language": "ja"}]}, "item_type_id": "10001", "owner": "15", "path": ["639"], "permalink_uri": "http://hdl.handle.net/10076/11104", "pubdate": {"attribute_name": "PubDate", "attribute_value": "2010-05-24"}, "publish_date": "2010-05-24", "publish_status": "0", "recid": "8383", "relation": {}, "relation_version_is_last": true, "title": ["ベイズアプローチによる最適識別系の有限標本効果に関する考察 : 学習標本の大きさがクラス間で異なる場合(\u003c論文小特集\u003eパターン認識のための学習 : 基礎と応用)"], "weko_shared_id": -1}