WEKO3
アイテム
{"_buckets": {"deposit": "d6e5bb39-3345-451b-ac25-8ea7b349aa3c"}, "_deposit": {"created_by": 2, "id": "13449", "owners": [2], "pid": {"revision_id": 0, "type": "depid", "value": "13449"}, "status": "published"}, "_oai": {"id": "oai:nagasaki-u.repo.nii.ac.jp:00013449", "sets": ["21"]}, "author_link": ["48639", "48640", "48637", "48638"], "item_2_biblio_info_6": {"attribute_name": "書誌情報", "attribute_value_mlt": [{"bibliographicIssueDates": {"bibliographicIssueDate": "2011", "bibliographicIssueDateType": "Issued"}, "bibliographicIssueNumber": "PART 1", "bibliographicPageEnd": "447", "bibliographicPageStart": "435", "bibliographicVolumeNumber": "6634 LNAI", "bibliographic_titles": [{"bibliographic_title": "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)"}]}]}, "item_2_description_4": {"attribute_name": "抄録", "attribute_value_mlt": [{"subitem_description": "This paper provides a new approach to topical trend analysis. Our aim is to improve the generalization power of latent Dirichlet allocation (LDA) by using document timestamps. Many previous works model topical trends by making latent topic distributions time-dependent. We propose a straightforward approach by preparing a different word multinomial distribution for each time point. Since this approach increases the number of parameters, overfitting becomes a critical issue. Our contribution to this issue is two-fold. First, we propose an effective way of defining Dirichlet priors over the word multinomials. Second, we propose a special scheduling of variational Bayesian (VB) inference. Comprehensive experiments with six datasets prove that our approach can improve LDA and also Topics over Time, a well-known variant of LDA, in terms of test data perplexity in the framework of VB inference.", "subitem_description_type": "Abstract"}]}, "item_2_description_63": {"attribute_name": "引用", "attribute_value_mlt": [{"subitem_description": "Lecture Notes in Computer Science, 6634 LNAI(PART 1), pp.435-447; 2011", "subitem_description_type": "Other"}]}, "item_2_publisher_33": {"attribute_name": "出版者", "attribute_value_mlt": [{"subitem_publisher": "Springer Verlag"}]}, "item_2_rights_13": {"attribute_name": "権利", "attribute_value_mlt": [{"subitem_rights": "© 2011 Springer-Verlag."}, {"subitem_rights": "The original publication is available at www.springerlink.com"}]}, "item_2_source_id_10": {"attribute_name": "書誌レコードID", "attribute_value_mlt": [{"subitem_source_identifier": "AA0071599X", "subitem_source_identifier_type": "NCID"}]}, "item_2_source_id_7": {"attribute_name": "ISSN", "attribute_value_mlt": [{"subitem_source_identifier": "03029743", "subitem_source_identifier_type": "ISSN"}]}, "item_2_source_id_8": {"attribute_name": "EISSN", "attribute_value_mlt": [{"subitem_source_identifier": "16113349", "subitem_source_identifier_type": "ISSN"}]}, "item_2_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": "48637", "nameIdentifierScheme": "WEKO"}]}, {"creatorNames": [{"creatorName": "Takasu, Atsuhiro"}], "nameIdentifiers": [{"nameIdentifier": "48638", "nameIdentifierScheme": "WEKO"}]}, {"creatorNames": [{"creatorName": "Shibata, Yuichiro"}], "nameIdentifiers": [{"nameIdentifier": "48639", "nameIdentifierScheme": "WEKO"}]}, {"creatorNames": [{"creatorName": "Oguri, Kiyoshi"}], "nameIdentifiers": [{"nameIdentifier": "48640", "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": "2020-12-22"}], "displaytype": "detail", "download_preview_message": "", "file_order": 0, "filename": "LNCS6634_435.pdf", "filesize": [{"value": "172.6 kB"}], "format": "application/pdf", "future_date_message": "", "is_thumbnail": false, "licensetype": "license_free", "mimetype": "application/pdf", "size": 172600.0, "url": {"label": "LNCS6634_435.pdf", "url": "https://nagasaki-u.repo.nii.ac.jp/record/13449/files/LNCS6634_435.pdf"}, "version_id": "669446e1-1a5e-4970-8c33-07613cb1590b"}]}, "item_keyword": {"attribute_name": "キーワード", "attribute_value_mlt": [{"subitem_subject": "Bayesian methods", "subitem_subject_scheme": "Other"}, {"subitem_subject": "parallelization", "subitem_subject_scheme": "Other"}, {"subitem_subject": "topic models", "subitem_subject_scheme": "Other"}, {"subitem_subject": "trend analysis", "subitem_subject_scheme": "Other"}, {"subitem_subject": "variational inference", "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": "Steering time-dependent estimation of posteriors with hyperparameter indexing in Bayesian topic models", "item_titles": {"attribute_name": "タイトル", "attribute_value_mlt": [{"subitem_title": "Steering time-dependent estimation of posteriors with hyperparameter indexing in Bayesian topic models"}]}, "item_type_id": "2", "owner": "2", "path": ["21"], "permalink_uri": "http://hdl.handle.net/10069/25516", "pubdate": {"attribute_name": "公開日", "attribute_value": "2011-08-11"}, "publish_date": "2011-08-11", "publish_status": "0", "recid": "13449", "relation": {}, "relation_version_is_last": true, "title": ["Steering time-dependent estimation of posteriors with hyperparameter indexing in Bayesian topic models"], "weko_shared_id": 2}
Steering time-dependent estimation of posteriors with hyperparameter indexing in Bayesian topic models
http://hdl.handle.net/10069/25516
http://hdl.handle.net/10069/2551676b39ff2-fde1-488b-a045-ed98e963b266
名前 / ファイル | ライセンス | アクション |
---|---|---|
LNCS6634_435.pdf (172.6 kB)
|
|
Item type | 学術雑誌論文 / Journal Article(1) | |||||
---|---|---|---|---|---|---|
公開日 | 2011-08-11 | |||||
タイトル | ||||||
タイトル | Steering time-dependent estimation of posteriors with hyperparameter indexing in Bayesian topic models | |||||
言語 | ||||||
言語 | jpn | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | Bayesian methods | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | parallelization | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | topic models | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | trend analysis | |||||
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_6501 | |||||
資源タイプ | journal article | |||||
著者 |
Masada, Tomonari
× Masada, Tomonari× Takasu, Atsuhiro× Shibata, Yuichiro× Oguri, Kiyoshi |
|||||
抄録 | ||||||
内容記述タイプ | Abstract | |||||
内容記述 | This paper provides a new approach to topical trend analysis. Our aim is to improve the generalization power of latent Dirichlet allocation (LDA) by using document timestamps. Many previous works model topical trends by making latent topic distributions time-dependent. We propose a straightforward approach by preparing a different word multinomial distribution for each time point. Since this approach increases the number of parameters, overfitting becomes a critical issue. Our contribution to this issue is two-fold. First, we propose an effective way of defining Dirichlet priors over the word multinomials. Second, we propose a special scheduling of variational Bayesian (VB) inference. Comprehensive experiments with six datasets prove that our approach can improve LDA and also Topics over Time, a well-known variant of LDA, in terms of test data perplexity in the framework of VB inference. | |||||
書誌情報 |
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 巻 6634 LNAI, 号 PART 1, p. 435-447, 発行日 2011 |
|||||
出版者 | ||||||
出版者 | Springer Verlag | |||||
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 | |||||
書誌レコードID | ||||||
収録物識別子タイプ | NCID | |||||
収録物識別子 | AA0071599X | |||||
権利 | ||||||
権利情報 | © 2011 Springer-Verlag. | |||||
権利 | ||||||
権利情報 | The original publication is available at www.springerlink.com | |||||
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 | |||||
引用 | ||||||
内容記述タイプ | Other | |||||
内容記述 | Lecture Notes in Computer Science, 6634 LNAI(PART 1), pp.435-447; 2011 |