Document-level event argument extraction method and device, equipment and medium
The invention provides a document-level event argument extraction method and device, equipment and a medium, and the method comprises the steps: inputting a document, obtaining an event set, a trigger word set, a role set and an argument set in the document, constructing an argument extraction stude...
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , |
|---|---|
| Format: | Patent |
| Sprache: | chi ; eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | |
|---|---|
| container_issue | |
| container_start_page | |
| container_title | |
| container_volume | |
| creator | LI LINGHAN LI HONGYANG ZHANG XINYU ZHANG RUOHAN HU YANLI CAO JIANWEI ZHOU ZHUANGYI LIU TENG WANG HAOYUAN TAN LIJUN SUN YINGJIE |
| description | The invention provides a document-level event argument extraction method and device, equipment and a medium, and the method comprises the steps: inputting a document, obtaining an event set, a trigger word set, a role set and an argument set in the document, constructing an argument extraction student model, and predicting the argument sets of all events based on the document. Establishing a teacher model of an event internal level for each event to obtain argument association in each event, and establishing a teacher model of an inter-event level to obtain cross-event argument association in the document, the method comprises the following steps: migrating event association knowledge to an argument extraction student model from a teacher model of an internal level of each event and a teacher model of an inter-event level on the basis of multi-teacher distillation by adopting a feature alignment and logit alignment mode, and then determining a total cost function of the argument extraction student model; and |
| format | Patent |
| fullrecord | <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_CN118673899A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>CN118673899A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_CN118673899A3</originalsourceid><addsrcrecordid>eNrjZAhwyU8uzU3NK9HNSS1LzVEAEnklColF6WBBhdSKkqLE5JLM_DyF3NSSjPwUhcS8FIWU1LLM5FQdhdTC0swCsDqQaG5qSmZpLg8Da1piTnEqL5TmZlB0cw1x9tBNLciPTy0uSExOzUstiXf2MzS0MDM3trC0dDQmRg0AvsQ3pQ</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Document-level event argument extraction method and device, equipment and medium</title><source>esp@cenet</source><creator>LI LINGHAN ; LI HONGYANG ; ZHANG XINYU ; ZHANG RUOHAN ; HU YANLI ; CAO JIANWEI ; ZHOU ZHUANGYI ; LIU TENG ; WANG HAOYUAN ; TAN LIJUN ; SUN YINGJIE</creator><creatorcontrib>LI LINGHAN ; LI HONGYANG ; ZHANG XINYU ; ZHANG RUOHAN ; HU YANLI ; CAO JIANWEI ; ZHOU ZHUANGYI ; LIU TENG ; WANG HAOYUAN ; TAN LIJUN ; SUN YINGJIE</creatorcontrib><description>The invention provides a document-level event argument extraction method and device, equipment and a medium, and the method comprises the steps: inputting a document, obtaining an event set, a trigger word set, a role set and an argument set in the document, constructing an argument extraction student model, and predicting the argument sets of all events based on the document. Establishing a teacher model of an event internal level for each event to obtain argument association in each event, and establishing a teacher model of an inter-event level to obtain cross-event argument association in the document, the method comprises the following steps: migrating event association knowledge to an argument extraction student model from a teacher model of an internal level of each event and a teacher model of an inter-event level on the basis of multi-teacher distillation by adopting a feature alignment and logit alignment mode, and then determining a total cost function of the argument extraction student model; and</description><language>chi ; eng</language><subject>CALCULATING ; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS ; COMPUTING ; COUNTING ; ELECTRIC DIGITAL DATA PROCESSING ; PHYSICS</subject><creationdate>2024</creationdate><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20240920&DB=EPODOC&CC=CN&NR=118673899A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,776,881,25542,76289</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20240920&DB=EPODOC&CC=CN&NR=118673899A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>LI LINGHAN</creatorcontrib><creatorcontrib>LI HONGYANG</creatorcontrib><creatorcontrib>ZHANG XINYU</creatorcontrib><creatorcontrib>ZHANG RUOHAN</creatorcontrib><creatorcontrib>HU YANLI</creatorcontrib><creatorcontrib>CAO JIANWEI</creatorcontrib><creatorcontrib>ZHOU ZHUANGYI</creatorcontrib><creatorcontrib>LIU TENG</creatorcontrib><creatorcontrib>WANG HAOYUAN</creatorcontrib><creatorcontrib>TAN LIJUN</creatorcontrib><creatorcontrib>SUN YINGJIE</creatorcontrib><title>Document-level event argument extraction method and device, equipment and medium</title><description>The invention provides a document-level event argument extraction method and device, equipment and a medium, and the method comprises the steps: inputting a document, obtaining an event set, a trigger word set, a role set and an argument set in the document, constructing an argument extraction student model, and predicting the argument sets of all events based on the document. Establishing a teacher model of an event internal level for each event to obtain argument association in each event, and establishing a teacher model of an inter-event level to obtain cross-event argument association in the document, the method comprises the following steps: migrating event association knowledge to an argument extraction student model from a teacher model of an internal level of each event and a teacher model of an inter-event level on the basis of multi-teacher distillation by adopting a feature alignment and logit alignment mode, and then determining a total cost function of the argument extraction student model; and</description><subject>CALCULATING</subject><subject>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</subject><subject>COMPUTING</subject><subject>COUNTING</subject><subject>ELECTRIC DIGITAL DATA PROCESSING</subject><subject>PHYSICS</subject><fulltext>true</fulltext><rsrctype>patent</rsrctype><creationdate>2024</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNrjZAhwyU8uzU3NK9HNSS1LzVEAEnklColF6WBBhdSKkqLE5JLM_DyF3NSSjPwUhcS8FIWU1LLM5FQdhdTC0swCsDqQaG5qSmZpLg8Da1piTnEqL5TmZlB0cw1x9tBNLciPTy0uSExOzUstiXf2MzS0MDM3trC0dDQmRg0AvsQ3pQ</recordid><startdate>20240920</startdate><enddate>20240920</enddate><creator>LI LINGHAN</creator><creator>LI HONGYANG</creator><creator>ZHANG XINYU</creator><creator>ZHANG RUOHAN</creator><creator>HU YANLI</creator><creator>CAO JIANWEI</creator><creator>ZHOU ZHUANGYI</creator><creator>LIU TENG</creator><creator>WANG HAOYUAN</creator><creator>TAN LIJUN</creator><creator>SUN YINGJIE</creator><scope>EVB</scope></search><sort><creationdate>20240920</creationdate><title>Document-level event argument extraction method and device, equipment and medium</title><author>LI LINGHAN ; LI HONGYANG ; ZHANG XINYU ; ZHANG RUOHAN ; HU YANLI ; CAO JIANWEI ; ZHOU ZHUANGYI ; LIU TENG ; WANG HAOYUAN ; TAN LIJUN ; SUN YINGJIE</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_CN118673899A3</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>chi ; eng</language><creationdate>2024</creationdate><topic>CALCULATING</topic><topic>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</topic><topic>COMPUTING</topic><topic>COUNTING</topic><topic>ELECTRIC DIGITAL DATA PROCESSING</topic><topic>PHYSICS</topic><toplevel>online_resources</toplevel><creatorcontrib>LI LINGHAN</creatorcontrib><creatorcontrib>LI HONGYANG</creatorcontrib><creatorcontrib>ZHANG XINYU</creatorcontrib><creatorcontrib>ZHANG RUOHAN</creatorcontrib><creatorcontrib>HU YANLI</creatorcontrib><creatorcontrib>CAO JIANWEI</creatorcontrib><creatorcontrib>ZHOU ZHUANGYI</creatorcontrib><creatorcontrib>LIU TENG</creatorcontrib><creatorcontrib>WANG HAOYUAN</creatorcontrib><creatorcontrib>TAN LIJUN</creatorcontrib><creatorcontrib>SUN YINGJIE</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>LI LINGHAN</au><au>LI HONGYANG</au><au>ZHANG XINYU</au><au>ZHANG RUOHAN</au><au>HU YANLI</au><au>CAO JIANWEI</au><au>ZHOU ZHUANGYI</au><au>LIU TENG</au><au>WANG HAOYUAN</au><au>TAN LIJUN</au><au>SUN YINGJIE</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Document-level event argument extraction method and device, equipment and medium</title><date>2024-09-20</date><risdate>2024</risdate><abstract>The invention provides a document-level event argument extraction method and device, equipment and a medium, and the method comprises the steps: inputting a document, obtaining an event set, a trigger word set, a role set and an argument set in the document, constructing an argument extraction student model, and predicting the argument sets of all events based on the document. Establishing a teacher model of an event internal level for each event to obtain argument association in each event, and establishing a teacher model of an inter-event level to obtain cross-event argument association in the document, the method comprises the following steps: migrating event association knowledge to an argument extraction student model from a teacher model of an internal level of each event and a teacher model of an inter-event level on the basis of multi-teacher distillation by adopting a feature alignment and logit alignment mode, and then determining a total cost function of the argument extraction student model; and</abstract><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | |
| ispartof | |
| issn | |
| language | chi ; eng |
| recordid | cdi_epo_espacenet_CN118673899A |
| source | esp@cenet |
| subjects | CALCULATING COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS COMPUTING COUNTING ELECTRIC DIGITAL DATA PROCESSING PHYSICS |
| title | Document-level event argument extraction method and device, equipment and medium |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-11T07%3A38%3A29IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-epo_EVB&rft_val_fmt=info:ofi/fmt:kev:mtx:patent&rft.genre=patent&rft.au=LI%20LINGHAN&rft.date=2024-09-20&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3ECN118673899A%3C/epo_EVB%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |