{"updated":"2025-01-20T04:54:43.674015+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00178980","sets":["1164:4088:9073:9141"]},"path":["9141"],"owner":"11","recid":"178980","title":["特徴抽出器の学習と購買履歴を必要としない類似画像による関連商品検索システム"],"pubdate":{"attribute_name":"公開日","attribute_value":"2017-05-18"},"_buckets":{"deposit":"84b5bbb2-4bd4-4f0b-bc4b-0097688d8583"},"_deposit":{"id":"178980","pid":{"type":"depid","value":"178980","revision_id":0},"owners":[11],"status":"published","created_by":11},"item_title":"特徴抽出器の学習と購買履歴を必要としない類似画像による関連商品検索システム","author_link":["384836","384835","384832","384837","384833","384838","384831","384834"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"特徴抽出器の学習と購買履歴を必要としない類似画像による関連商品検索システム"},{"subitem_title":"A search system of retrieving images of the similar products without the requirement of the training of feature extractor and the purchase history","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"システム構築,プライバシー保護技術","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2017-05-18","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"GMOペパボ株式会社ペパボ研究所"},{"subitem_text_value":"GMOペパボ株式会社ペパボ研究所"},{"subitem_text_value":"GMOペパボ株式会社ペパボ研究所/力武健次技術士事務所"},{"subitem_text_value":"GMOペパボ株式会社ペパボ研究所"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Pepabo R&D Institute, GMO Pepabo, Inc.","subitem_text_language":"en"},{"subitem_text_value":"Pepabo R&D Institute, GMO Pepabo, Inc.","subitem_text_language":"en"},{"subitem_text_value":"Pepabo R&D Institute, GMO Pepabo, Inc. / Kenji Rikitake Professional Engineer's Office","subitem_text_language":"en"},{"subitem_text_value":"Pepabo R&D Institute, GMO Pepabo, Inc.","subitem_text_language":"en"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_publisher":{"attribute_name":"出版者","attribute_value_mlt":[{"subitem_publisher":"情報処理学会","subitem_publisher_language":"ja"}]},"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/178980/files/IPSJ-IOT17037004.pdf","label":"IPSJ-IOT17037004.pdf"},"date":[{"dateType":"Available","dateValue":"2019-05-18"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-IOT17037004.pdf","filesize":[{"value":"1.6 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":"43"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"2ca7cf39-d738-4627-8bbc-69363a324321","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2017 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"三宅, 悠介"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"松本, 亮介"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"力武, 健次"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"栗林, 健太郎"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Yusuke, Miyake","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Ryosuke, Matsumoto","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Kenji, Rikitake","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Kentaro, Kuribayashi","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AA12326962","subitem_source_identifier_type":"NCID"}]},"item_4_textarea_12":{"attribute_name":"Notice","attribute_value_mlt":[{"subitem_textarea_value":"SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc."}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_18gh","resourcetype":"technical report"}]},"item_4_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"2188-8787","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"BtoC の EC サイトで取り扱う商品の種類の増加に伴い,EC サイト利用者の通常の行動では全ての商品を見て回ることは困難であるため,多くの EC サイトでは効率的に商品を閲覧できるよう関連性のある商品を動線上に表示している.購買履歴等の情報が蓄積されないと関連商品を選定できない問題を解決するため,商品の持つ様々なメタデータを利用する手法や,視覚的な訴求力の強い商品画像を元にした,畳み込みニューラルネットワークを始めとした深層学習による精度の高い関連商品の選定手法が提案されている.しかし,適切な粒度のメタデータの整備に手間を要する問題や,深層学習のための大量の訓練データセットと計算時間が必要となる問題から,これらが導入への大きな障壁となっている.本報告では,画像分類用の学術ベンチマークであり,EC サイト商品画像特性と類似する ImageNet において高い成績を出した Inception-v3 モデルを学習済みネットワークとして採用し,一般物体の特徴を強く表現する識別層に近い手前のプーリング層までから得られる特徴量をもとに近似最近傍探索により類似度を比較することで,特徴抽出器の学習と購買履歴を必要としない類似画像による関連商品検索システムを提案する.EC サイトにこの類似画像による関連商品検索システムを導入し,画像のクリック率を商品カテゴリごとに計測することで類似画像による関連商品の有効性を検証した.","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Electronic Commerce (EC) sites show items of the user's interest following the flow for efficient browsing of the products. For enabling to choose the related products without the activity record of the users, methods using the metadata and the images of the products to suggest a highly accurate choice by deep learning with convolution neural networks have been proposed. Those methods, however, require the large amount of training data and calculation time with a properly structured product metadata, which results in the impediment to the production system deployment. In this report, we propose a search system of retrieving images of the similar products without the requirement of the training of feature extractor and the purchase history, by comparing the similarity with the approximate nearest neighbor search based on the features from the pooling layer before the identification layer, using the Inception-v3 model which claims a good result on ImageNet, an academic benchmark of image classification similar to that of EC site product image characteristics. We implement this system to an EC site and measure the clickthrough rate of the image for each product category to evaluate the effectiveness of directing the user flow by showing the related products by similar images.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"8","bibliographic_titles":[{"bibliographic_title":"研究報告インターネットと運用技術(IOT)"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2017-05-18","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"4","bibliographicVolumeNumber":"2017-IOT-37"}]},"relation_version_is_last":true,"weko_creator_id":"11"},"created":"2025-01-19T00:48:16.116789+00:00","id":178980,"links":{}}