{"created":"2023-05-15T08:44:02.950629+00:00","id":9265,"links":{},"metadata":{"_buckets":{"deposit":"2ff86225-d3e2-4c7c-b919-261d977acad3"},"_deposit":{"created_by":13,"id":"9265","owners":[13],"pid":{"revision_id":0,"type":"depid","value":"9265"},"status":"published"},"_oai":{"id":"oai:uec.repo.nii.ac.jp:00009265","sets":["6"]},"author_link":["5149","25364"],"control_number":"9265","item_10001_biblio_info_7":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicIssueDates":{"bibliographicIssueDate":"2016-11-15","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"11","bibliographicPageEnd":"2435","bibliographicPageStart":"2426","bibliographicVolumeNumber":"57","bibliographic_titles":[{"bibliographic_title":"情報処理学会論文誌","bibliographic_titleLang":"ja"}]}]},"item_10001_description_5":{"attribute_name":"抄録","attribute_value_mlt":[{"subitem_description":"ゲームの熟達支援システムの実現において，対局中にプレイヤの技能に合わせて不自然でなく弱く指すAIが重要な役割を持っている．しかしながら，プレイヤの技能に動的に合わせる将棋AIは実現しておらず，弱く調整した将棋AIに生じる不自然さも問題とされながら定量的には評価されていない．そこで本研究では，不自然でなく動的に弱く指す将棋AIの開発を目標に，ゲーム木探索によって局面の評価値を0に近づける着手を選択する手法を提案した．さらに，この提案手法を導入したAIを3つの実験から評価した．はじめに，探索の浅いAIとの対局実験では提案AIの勝率を調整する能力を評価した．次に，初心者と経験者による対局実験では，実験参加者に思考を発話させながらすべてのAIの着手を評価させ，提案手法を導入したAIとそうでないAIでの悪手検出率を比較した．最後に，インターネット上での対局実験では，提案AIによって強さを合わせられるプレイヤの幅を分析した．これらの実験の結果，初心者から級位者の勝率と強さの主観評価において，提案AIは適度な強さだと示された．さらに提案AIは既存AIと比較して悪手を多く生成するが，その悪手検出率は初心者の場合には5%を下回ることが分かった．","subitem_description_type":"Abstract"},{"subitem_description":"Game playing AI which can adjust its skill level dynamically for each opponent player without artificiality has important role in implementing computer-assisted instruction system for game playing. However, the existing methods cannot adjust the skill level of a Shogi program dynamically. In addition to this problem, artificiality with Shogi program made for weak players has not been measured quantitatively. Therefore, the present paper reports on a novel method for dynamic adjustment of the skill level of a Shogi program for each opponent player without artificiality. This method calculates and selects a move which approximates evaluation value of position to zero by game tree search and evaluation value function. Furthermore, this proposed method was evaluated by three experiments. First, ability for winning rate adjustment of the proposed method was evaluated by matches with weak game playing AI. Second, think aloud data and the subjective evaluations of all moves were gathered by matches with novice and person with experience. We compared detection rates of the proposed and existing AI. Finally, the range of players whom strength was matched with proposed AI was confirmed by matches on internet Shogi server. From results of these experiments, proposed AI was shown as proper strength for novice to kyu-players by winning rate and subjective evaluation. In addition, proposed AI makes bad moves more than existing AI, nevertheless, its detection rate of bad moves were lower than 5% for novice players.","subitem_description_type":"Abstract"}]},"item_10001_publisher_8":{"attribute_name":"出版者","attribute_value_mlt":[{"subitem_publisher":"情報処理学会"}]},"item_10001_relation_17":{"attribute_name":"関連サイト","attribute_value_mlt":[{"subitem_relation_type_id":{"subitem_relation_type_id_text":"http://id.nii.ac.jp/1001/00175926/","subitem_relation_type_select":"URI"}}]},"item_10001_rights_15":{"attribute_name":"権利","attribute_value_mlt":[{"subitem_rights":"(c) 2016 Information Processing Society of Japan. 本著作物の著作権は情報処理学会に帰属します。本著作物は著作権者である情報処理学会の許可のもとに掲載するものです。ご利用に当たっては「著作権法」ならびに「情報処理学会倫理綱領」に従うことをお願いいたします。"}]},"item_10001_source_id_9":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"1882-7764","subitem_source_identifier_type":"ISSN"}]},"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":"仲道, 隆史","creatorNameLang":"ja"},{"creatorName":"ナカミチ, タカフミ","creatorNameLang":"ja-Kana"},{"creatorName":"Nakamichi, Takafumi","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"伊藤, 毅志","creatorNameLang":"ja"},{"creatorName":"イトウ, タケシ","creatorNameLang":"ja-Kana"},{"creatorName":"Ito, Takeshi","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_files":{"attribute_name":"ファイル情報","attribute_type":"file","attribute_value_mlt":[{"accessrole":"open_date","date":[{"dateType":"Available","dateValue":"2019-06-20"}],"displaytype":"detail","filename":"IPSJ-JNL5711011.pdf","filesize":[{"value":"952.1 kB"}],"format":"application/pdf","licensetype":"license_note","mimetype":"application/pdf","url":{"label":"IPSJ-JNL5711011","url":"https://uec.repo.nii.ac.jp/record/9265/files/IPSJ-JNL5711011.pdf"},"version_id":"2ab52044-51b3-4adf-afb3-11fb48a02856"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"コンピュータ将棋","subitem_subject_scheme":"Other"},{"subitem_subject":"学習支援","subitem_subject_scheme":"Other"},{"subitem_subject":"自然に弱く指すAI","subitem_subject_scheme":"Other"},{"subitem_subject":"人間らしさ","subitem_subject_scheme":"Other"},{"subitem_subject":"指導将棋AI","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":"プレイヤの技能に動的に合わせるシステムの提案と評価","item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"プレイヤの技能に動的に合わせるシステムの提案と評価","subitem_title_language":"ja"},{"subitem_title":"Proposal and Evaluation of System of Dynamic Adapting Method to Player's Skill","subitem_title_language":"en"}]},"item_type_id":"10001","owner":"13","path":["6"],"pubdate":{"attribute_name":"PubDate","attribute_value":"2019-06-20"},"publish_date":"2019-06-20","publish_status":"0","recid":"9265","relation_version_is_last":true,"title":["プレイヤの技能に動的に合わせるシステムの提案と評価"],"weko_creator_id":"13","weko_shared_id":-1},"updated":"2024-03-04T07:29:26.926330+00:00"}