{"created":"2023-05-15T14:21:31.065918+00:00","id":3197,"links":{},"metadata":{"_buckets":{"deposit":"415a81fe-82dc-4314-92d0-d90f71991046"},"_deposit":{"created_by":3,"id":"3197","owners":[3],"pid":{"revision_id":0,"type":"depid","value":"3197"},"status":"published"},"_oai":{"id":"oai:bunkyo.repo.nii.ac.jp:00003197","sets":["1:26:203"]},"author_link":["4140"],"item_5_biblio_info_13":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicIssueDates":{"bibliographicIssueDate":"1998-01-01"},"bibliographicPageEnd":"119","bibliographicPageStart":"83","bibliographicVolumeNumber":"19","bibliographic_titles":[{"bibliographic_title":"情報研究"},{"bibliographic_title":"Information and Communication Studies"}]}]},"item_5_date_43":{"attribute_name":"作成日","attribute_value_mlt":[{"subitem_date_issued_datetime":"2011-02-22"}]},"item_5_description_12":{"attribute_name":"抄録","attribute_value_mlt":[{"subitem_description":"　不動点探索形構造受精に関する多段階帰納推論を使ったパターン認識アルゴリズムを内蔵している認識システムRECOGNITRONにおいては、モデル構成作用素T、類似度関数SM、大分類関数BSCを各変換段階で用い、入力パターンの帰属するカテゴリを仮決定しながら、次の段階へと進むとき、訂正できる機能がある多段階構成がとられている。この多段階構成アルゴリズムは、処理の対象とする問題のパターンφについてのパターンモデルの列(知覚的記憶表彰の列)を再生しながら、\"パターンのモデルとその帰属する可能性のあるカテゴリの番号のリストとの対として定義されるカテゴリ帰属知識\"に関する構造受精変換の不動点として、最終的には、\"その帰属するカテゴリの代表パターンのモデルとそのカテゴリ番号のみからなるリストとの対\"を確保する認識手法である。\n　処理の対象としたパターンφの有限部分集合Φ?(⊆Φ)に対する認識率が良好でない場合、その原因として、3構成要素T,SM,BSCがΦ?に関し、適切に選定されていなかったことが先ず、挙げられる。\n　本論文の主目的は、T,BSCが適切に選ばれている状況の下で、有限部分パターン集合Φ?について、SMが適切に選定されているかどうかを赤池情報量基準AICを用いて判定する方法を、数理的に研究することである。本認識アルゴリズムがSS公理系で構築されており、従って、axiom 2を満たすSMが多数存在することを考慮すると、選ばれた1つのSMが適切かどかを検証するこのような方法は必要なものである。\n\\n　The recognition system RECOGNITRON having a pattern-recognition algorithm of a multistage induction inference using structural fertilization transformation seems to seek for highly probable categories at a new stage determining temporary categories of an input pattern φ in question at an old stage. A given input pattern φ is transformed into a sequence of categorical membership-knowleges which is defined as a pair of a pattern-model and a list of its category-numbers obtained at each stage using model-construction operator T, similarity-measure function SM, and rough classifiers BSC, and at the final stage is reproduced as the first half(the model corresponding to the prototypical pattern Tωj of the j-category ?j to which φ may belong) of the fixed-point knowledge of a selected structural-fertilization transformation.\n　If a probability of misrecognition is low for a finite subset Φ?⊆Φ (a set of patterns to be recognized), three fundamental constituents T, SM and BSC of RECOGNITRON are measures inadequate to the situation Φ?.\n　The main purpose of this paper is to mathematically investigate a method of testing whether or not using AIC(the Akaike information criterion) SM is adequate to Φ? on the assumption that T and BSC were selected adequately. In consideration of that the recognition algorithm holds good under a system of SS-axioms and therefore there are many SMs satisfying axiom 2 of SS-axioms, such a method of testing the selected SM for a adequacy is necessary.","subitem_description_type":"Abstract"}]},"item_5_description_38":{"attribute_name":"フォーマット","attribute_value_mlt":[{"subitem_description":"application/pdf","subitem_description_type":"Other"}]},"item_5_source_id_19":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"03893367"}]},"item_5_text_39":{"attribute_name":"本文言語","attribute_value_mlt":[{"subitem_text_value":"日本語"}]},"item_5_text_42":{"attribute_name":"ID","attribute_value_mlt":[{"subitem_text_value":"BKSJ190005"}]},"item_5_text_7":{"attribute_name":"Author","attribute_value_mlt":[{"subitem_text_value":"Suzuki, Shoichi"}]},"item_5_text_8":{"attribute_name":"所属機関","attribute_value_mlt":[{"subitem_text_value":"文教大学情報学部"}]},"item_5_text_9":{"attribute_name":"Institution","attribute_value_mlt":[{"subitem_text_value":"Bunkyo University Faculty of Information and Communications"}]},"item_5_version_type_35":{"attribute_name":"著者版フラグ","attribute_value_mlt":[{"subitem_version_type":"VoR"}]},"item_creator":{"attribute_name":"著者","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"鈴木, 昇一"}],"nameIdentifiers":[{}]}]},"item_files":{"attribute_name":"ファイル情報","attribute_type":"file","attribute_value_mlt":[{"accessrole":"open_date","date":[{"dateType":"Available","dateValue":"2018-03-24"}],"displaytype":"detail","filename":"BKSJ190005.pdf","filesize":[{"value":"2.1 MB"}],"format":"application/pdf","licensetype":"license_note","mimetype":"application/pdf","url":{"label":"BKSJ190005.pdf","url":"https://bunkyo.repo.nii.ac.jp/record/3197/files/BKSJ190005.pdf"},"version_id":"19d19e6f-078c-4d6b-b414-454bd59af387"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"モデル構成作用素"},{"subitem_subject":"類似度関数"},{"subitem_subject":"大分類関数"},{"subitem_subject":"カテゴリ帰属知識の不動点"},{"subitem_subject":"構造受精変換"},{"subitem_subject":"適合度検定"},{"subitem_subject":"X?確率分布"},{"subitem_subject":"赤池情報量基準"}]},"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":"departmental bulletin paper","resourceuri":"http://purl.org/coar/resource_type/c_6501"}]},"item_title":"類似度関数の選定に関する適切さの検証法","item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"類似度関数の選定に関する適切さの検証法"},{"subitem_title":"A Method for a Test of a Statistical Hypothesis about Whether or Not a Selected Similarity-Measure is Proper"}]},"item_type_id":"5","owner":"3","path":["203"],"pubdate":{"attribute_name":"公開日","attribute_value":"2011-02-22"},"publish_date":"2011-02-22","publish_status":"0","recid":"3197","relation_version_is_last":true,"title":["類似度関数の選定に関する適切さの検証法"],"weko_creator_id":"3","weko_shared_id":-1},"updated":"2023-05-16T16:47:03.477559+00:00"}