WEKO3
アイテム
{"_buckets": {"deposit": "9a16d7e1-72c9-43c0-bb8a-70683504a633"}, "_deposit": {"created_by": 2, "id": "6724", "owners": [2], "pid": {"revision_id": 0, "type": "depid", "value": "6724"}, "status": "published"}, "_oai": {"id": "oai:nagasaki-u.repo.nii.ac.jp:00006724", "sets": ["65"]}, "author_link": ["28279", "28278"], "item_9_biblio_info_6": {"attribute_name": "書誌情報", "attribute_value_mlt": [{"bibliographicIssueDates": {"bibliographicIssueDate": "2013-07", "bibliographicIssueDateType": "Issued"}, "bibliographicPageEnd": "454", "bibliographicPageStart": "445", "bibliographicVolumeNumber": "7952", "bibliographic_titles": [{"bibliographic_title": "Lecture Notes in Computer Science"}]}]}, "item_9_description_4": {"attribute_name": "抄録", "attribute_value_mlt": [{"subitem_description": "In this paper, we provide a revised inference for correlated topic model (CTM) [3]. CTM is proposed by Blei et al. for modeling correlations among latent topics more expressively than latent Dirichlet allocation (LDA) [2] and has been attracting attention of researchers. However, we have found that the variational inference of the original paper is unstable due to almost-singularity of the covariance matrix when the number of topics is large. This means that we may be reluctant to use CTM for analyzing a large document set, which may cover a rich diversity of topics. Therefore, we revise the inference and improve its quality. First, we modify the formula for updating the covariance matrix in a manner that enables us to recover the original inference by adjusting a parameter. Second, we regularize posterior parameters for reducing a side effect caused by the formula modification. While our method is based on a heuristic intuition, an experiment conducted on large document sets showed that it worked effectively in terms of perplexity.", "subitem_description_type": "Abstract"}]}, "item_9_description_5": {"attribute_name": "内容記述", "attribute_value_mlt": [{"subitem_description": "10th International Symposium on Neural Networks, ISNN 2013; Dalian; China; 4 July 2013 through 6 July 2013", "subitem_description_type": "Other"}]}, "item_9_description_63": {"attribute_name": "引用", "attribute_value_mlt": [{"subitem_description": "Lecture Notes in Computer Science, 7952, pp.445-454; 2013", "subitem_description_type": "Other"}]}, "item_9_publisher_33": {"attribute_name": "出版者", "attribute_value_mlt": [{"subitem_publisher": "Springer Verlag"}]}, "item_9_relation_12": {"attribute_name": "DOI", "attribute_value_mlt": [{"subitem_relation_type": "isVersionOf", "subitem_relation_type_id": {"subitem_relation_type_id_text": "10.1007/978-3-642-39068-5_54", "subitem_relation_type_select": "DOI"}}]}, "item_9_rights_13": {"attribute_name": "権利", "attribute_value_mlt": [{"subitem_rights": "© 2013 Springer-Verlag Berlin Heidelberg."}]}, "item_9_source_id_7": {"attribute_name": "ISSN", "attribute_value_mlt": [{"subitem_source_identifier": "03029743", "subitem_source_identifier_type": "ISSN"}]}, "item_9_source_id_8": {"attribute_name": "EISSN", "attribute_value_mlt": [{"subitem_source_identifier": "16113349", "subitem_source_identifier_type": "ISSN"}]}, "item_9_version_type_16": {"attribute_name": "著者版フラグ", "attribute_value_mlt": [{"subitem_version_resource": "http://purl.org/coar/version/c_ab4af688f83e57aa", "subitem_version_type": "AM"}]}, "item_creator": {"attribute_name": "著者", "attribute_type": "creator", "attribute_value_mlt": [{"creatorNames": [{"creatorName": "Masada, Tomonari"}], "nameIdentifiers": [{"nameIdentifier": "28278", "nameIdentifierScheme": "WEKO"}]}, {"creatorNames": [{"creatorName": "Takasu, Atsuhiro"}], "nameIdentifiers": [{"nameIdentifier": "28279", "nameIdentifierScheme": "WEKO"}]}]}, "item_files": {"attribute_name": "ファイル情報", "attribute_type": "file", "attribute_value_mlt": [{"accessrole": "open_date", "date": [{"dateType": "Available", "dateValue": "2020-12-21"}], "displaytype": "detail", "download_preview_message": "", "file_order": 0, "filename": "LNCS7952_445.pdf", "filesize": [{"value": "1.8 MB"}], "format": "application/pdf", "future_date_message": "", "is_thumbnail": false, "licensetype": "license_free", "mimetype": "application/pdf", "size": 1800000.0, "url": {"label": "LNCS7952_445.pdf", "url": "https://nagasaki-u.repo.nii.ac.jp/record/6724/files/LNCS7952_445.pdf"}, "version_id": "4aef9a23-eb7e-4754-8923-c42aec43b5e5"}]}, "item_keyword": {"attribute_name": "キーワード", "attribute_value_mlt": [{"subitem_subject": "covariance matrix", "subitem_subject_scheme": "Other"}, {"subitem_subject": "Topic models", "subitem_subject_scheme": "Other"}, {"subitem_subject": "variational inference", "subitem_subject_scheme": "Other"}]}, "item_language": {"attribute_name": "言語", "attribute_value_mlt": [{"subitem_language": "eng"}]}, "item_resource_type": {"attribute_name": "資源タイプ", "attribute_value_mlt": [{"resourcetype": "conference paper", "resourceuri": "http://purl.org/coar/resource_type/c_5794"}]}, "item_title": "A Revised Inference for Correlated Topic Model", "item_titles": {"attribute_name": "タイトル", "attribute_value_mlt": [{"subitem_title": "A Revised Inference for Correlated Topic Model"}]}, "item_type_id": "9", "owner": "2", "path": ["65"], "permalink_uri": "http://hdl.handle.net/10069/33725", "pubdate": {"attribute_name": "公開日", "attribute_value": "2014-08-01"}, "publish_date": "2014-08-01", "publish_status": "0", "recid": "6724", "relation": {}, "relation_version_is_last": true, "title": ["A Revised Inference for Correlated Topic Model"], "weko_shared_id": -1}
A Revised Inference for Correlated Topic Model
http://hdl.handle.net/10069/33725
http://hdl.handle.net/10069/33725e1458166-38ed-4b95-99b3-afb0e431b3c3
名前 / ファイル | ライセンス | アクション |
---|---|---|
LNCS7952_445.pdf (1.8 MB)
|
|
Item type | 会議発表論文 / Conference Paper(1) | |||||
---|---|---|---|---|---|---|
公開日 | 2014-08-01 | |||||
タイトル | ||||||
タイトル | A Revised Inference for Correlated Topic Model | |||||
言語 | ||||||
言語 | eng | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | covariance matrix | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | Topic models | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | variational inference | |||||
資源タイプ | ||||||
資源タイプ識別子 | http://purl.org/coar/resource_type/c_5794 | |||||
資源タイプ | conference paper | |||||
著者 |
Masada, Tomonari
× Masada, Tomonari× Takasu, Atsuhiro |
|||||
抄録 | ||||||
内容記述タイプ | Abstract | |||||
内容記述 | In this paper, we provide a revised inference for correlated topic model (CTM) [3]. CTM is proposed by Blei et al. for modeling correlations among latent topics more expressively than latent Dirichlet allocation (LDA) [2] and has been attracting attention of researchers. However, we have found that the variational inference of the original paper is unstable due to almost-singularity of the covariance matrix when the number of topics is large. This means that we may be reluctant to use CTM for analyzing a large document set, which may cover a rich diversity of topics. Therefore, we revise the inference and improve its quality. First, we modify the formula for updating the covariance matrix in a manner that enables us to recover the original inference by adjusting a parameter. Second, we regularize posterior parameters for reducing a side effect caused by the formula modification. While our method is based on a heuristic intuition, an experiment conducted on large document sets showed that it worked effectively in terms of perplexity. | |||||
内容記述 | ||||||
内容記述タイプ | Other | |||||
内容記述 | 10th International Symposium on Neural Networks, ISNN 2013; Dalian; China; 4 July 2013 through 6 July 2013 | |||||
書誌情報 |
Lecture Notes in Computer Science 巻 7952, p. 445-454, 発行日 2013-07 |
|||||
ISSN | ||||||
収録物識別子タイプ | ISSN | |||||
収録物識別子 | 03029743 | |||||
EISSN | ||||||
収録物識別子タイプ | ISSN | |||||
収録物識別子 | 16113349 | |||||
DOI | ||||||
関連タイプ | isVersionOf | |||||
識別子タイプ | DOI | |||||
関連識別子 | 10.1007/978-3-642-39068-5_54 | |||||
権利 | ||||||
権利情報 | © 2013 Springer-Verlag Berlin Heidelberg. | |||||
著者版フラグ | ||||||
出版タイプ | AM | |||||
出版タイプResource | http://purl.org/coar/version/c_ab4af688f83e57aa | |||||
出版者 | ||||||
出版者 | Springer Verlag | |||||
引用 | ||||||
内容記述タイプ | Other | |||||
内容記述 | Lecture Notes in Computer Science, 7952, pp.445-454; 2013 |