Text matching method based on data enhancement and graph matching network
The invention discloses a text matching method based on data enhancement and a graph matching network, which introduces dependency syntactic analysis and establishes a dependency relationship between text semantic units. And focusing the model on the core semantic unit by using a self-attention mech...
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 | LIU NING ZHAN DAIYI ZHANG SIQI LIU JINSHUO DENG JUAN TANG HAOZHOU WANG CHENYANG LIU KAI HUANG SHUO |
| description | The invention discloses a text matching method based on data enhancement and a graph matching network, which introduces dependency syntactic analysis and establishes a dependency relationship between text semantic units. And focusing the model on the core semantic unit by using a self-attention mechanism. Through a graph matching network, connection is established between any words of two text segments, full interaction is carried out through an attention mechanism, semantic focuses are better grasped, and the similarity between the texts is learned. And realizing data enhancement of the question matching data set from word and sentence granularity. An entity replacement algorithm, a synonym replacement and random insertion algorithm, a word noise enhancement algorithm and a back translation algorithm are designed, so that the diversity of a data set is improved, and the problems of insufficient corpora and the like are solved. According to the method, dependency syntactic analysis and a graph matching networ |
| format | Patent |
| fullrecord | <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_CN115510841A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>CN115510841A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_CN115510841A3</originalsourceid><addsrcrecordid>eNrjZPAMSa0oUchNLEnOyMxLV8hNLcnIT1FISixOTVHIz1NISSxJVEjNy0jMS07NTc0rUUjMS1FIL0osyEDoyUstKc8vyuZhYE1LzClO5YXS3AyKbq4hzh66qQX58anFBYnJqUCV8c5-hoampoYGFiaGjsbEqAEAvn40aw</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Text matching method based on data enhancement and graph matching network</title><source>esp@cenet</source><creator>LIU NING ; ZHAN DAIYI ; ZHANG SIQI ; LIU JINSHUO ; DENG JUAN ; TANG HAOZHOU ; WANG CHENYANG ; LIU KAI ; HUANG SHUO</creator><creatorcontrib>LIU NING ; ZHAN DAIYI ; ZHANG SIQI ; LIU JINSHUO ; DENG JUAN ; TANG HAOZHOU ; WANG CHENYANG ; LIU KAI ; HUANG SHUO</creatorcontrib><description>The invention discloses a text matching method based on data enhancement and a graph matching network, which introduces dependency syntactic analysis and establishes a dependency relationship between text semantic units. And focusing the model on the core semantic unit by using a self-attention mechanism. Through a graph matching network, connection is established between any words of two text segments, full interaction is carried out through an attention mechanism, semantic focuses are better grasped, and the similarity between the texts is learned. And realizing data enhancement of the question matching data set from word and sentence granularity. An entity replacement algorithm, a synonym replacement and random insertion algorithm, a word noise enhancement algorithm and a back translation algorithm are designed, so that the diversity of a data set is improved, and the problems of insufficient corpora and the like are solved. According to the method, dependency syntactic analysis and a graph matching networ</description><language>chi ; eng</language><subject>CALCULATING ; COMPUTING ; COUNTING ; ELECTRIC DIGITAL DATA PROCESSING ; HANDLING RECORD CARRIERS ; PHYSICS ; PRESENTATION OF DATA ; RECOGNITION OF DATA ; RECORD CARRIERS</subject><creationdate>2022</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=20221223&DB=EPODOC&CC=CN&NR=115510841A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,780,885,25563,76318</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20221223&DB=EPODOC&CC=CN&NR=115510841A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>LIU NING</creatorcontrib><creatorcontrib>ZHAN DAIYI</creatorcontrib><creatorcontrib>ZHANG SIQI</creatorcontrib><creatorcontrib>LIU JINSHUO</creatorcontrib><creatorcontrib>DENG JUAN</creatorcontrib><creatorcontrib>TANG HAOZHOU</creatorcontrib><creatorcontrib>WANG CHENYANG</creatorcontrib><creatorcontrib>LIU KAI</creatorcontrib><creatorcontrib>HUANG SHUO</creatorcontrib><title>Text matching method based on data enhancement and graph matching network</title><description>The invention discloses a text matching method based on data enhancement and a graph matching network, which introduces dependency syntactic analysis and establishes a dependency relationship between text semantic units. And focusing the model on the core semantic unit by using a self-attention mechanism. Through a graph matching network, connection is established between any words of two text segments, full interaction is carried out through an attention mechanism, semantic focuses are better grasped, and the similarity between the texts is learned. And realizing data enhancement of the question matching data set from word and sentence granularity. An entity replacement algorithm, a synonym replacement and random insertion algorithm, a word noise enhancement algorithm and a back translation algorithm are designed, so that the diversity of a data set is improved, and the problems of insufficient corpora and the like are solved. According to the method, dependency syntactic analysis and a graph matching networ</description><subject>CALCULATING</subject><subject>COMPUTING</subject><subject>COUNTING</subject><subject>ELECTRIC DIGITAL DATA PROCESSING</subject><subject>HANDLING RECORD CARRIERS</subject><subject>PHYSICS</subject><subject>PRESENTATION OF DATA</subject><subject>RECOGNITION OF DATA</subject><subject>RECORD CARRIERS</subject><fulltext>true</fulltext><rsrctype>patent</rsrctype><creationdate>2022</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNrjZPAMSa0oUchNLEnOyMxLV8hNLcnIT1FISixOTVHIz1NISSxJVEjNy0jMS07NTc0rUUjMS1FIL0osyEDoyUstKc8vyuZhYE1LzClO5YXS3AyKbq4hzh66qQX58anFBYnJqUCV8c5-hoampoYGFiaGjsbEqAEAvn40aw</recordid><startdate>20221223</startdate><enddate>20221223</enddate><creator>LIU NING</creator><creator>ZHAN DAIYI</creator><creator>ZHANG SIQI</creator><creator>LIU JINSHUO</creator><creator>DENG JUAN</creator><creator>TANG HAOZHOU</creator><creator>WANG CHENYANG</creator><creator>LIU KAI</creator><creator>HUANG SHUO</creator><scope>EVB</scope></search><sort><creationdate>20221223</creationdate><title>Text matching method based on data enhancement and graph matching network</title><author>LIU NING ; ZHAN DAIYI ; ZHANG SIQI ; LIU JINSHUO ; DENG JUAN ; TANG HAOZHOU ; WANG CHENYANG ; LIU KAI ; HUANG SHUO</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_CN115510841A3</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>chi ; eng</language><creationdate>2022</creationdate><topic>CALCULATING</topic><topic>COMPUTING</topic><topic>COUNTING</topic><topic>ELECTRIC DIGITAL DATA PROCESSING</topic><topic>HANDLING RECORD CARRIERS</topic><topic>PHYSICS</topic><topic>PRESENTATION OF DATA</topic><topic>RECOGNITION OF DATA</topic><topic>RECORD CARRIERS</topic><toplevel>online_resources</toplevel><creatorcontrib>LIU NING</creatorcontrib><creatorcontrib>ZHAN DAIYI</creatorcontrib><creatorcontrib>ZHANG SIQI</creatorcontrib><creatorcontrib>LIU JINSHUO</creatorcontrib><creatorcontrib>DENG JUAN</creatorcontrib><creatorcontrib>TANG HAOZHOU</creatorcontrib><creatorcontrib>WANG CHENYANG</creatorcontrib><creatorcontrib>LIU KAI</creatorcontrib><creatorcontrib>HUANG SHUO</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>LIU NING</au><au>ZHAN DAIYI</au><au>ZHANG SIQI</au><au>LIU JINSHUO</au><au>DENG JUAN</au><au>TANG HAOZHOU</au><au>WANG CHENYANG</au><au>LIU KAI</au><au>HUANG SHUO</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Text matching method based on data enhancement and graph matching network</title><date>2022-12-23</date><risdate>2022</risdate><abstract>The invention discloses a text matching method based on data enhancement and a graph matching network, which introduces dependency syntactic analysis and establishes a dependency relationship between text semantic units. And focusing the model on the core semantic unit by using a self-attention mechanism. Through a graph matching network, connection is established between any words of two text segments, full interaction is carried out through an attention mechanism, semantic focuses are better grasped, and the similarity between the texts is learned. And realizing data enhancement of the question matching data set from word and sentence granularity. An entity replacement algorithm, a synonym replacement and random insertion algorithm, a word noise enhancement algorithm and a back translation algorithm are designed, so that the diversity of a data set is improved, and the problems of insufficient corpora and the like are solved. According to the method, dependency syntactic analysis and a graph matching networ</abstract><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | |
| ispartof | |
| issn | |
| language | chi ; eng |
| recordid | cdi_epo_espacenet_CN115510841A |
| source | esp@cenet |
| subjects | CALCULATING COMPUTING COUNTING ELECTRIC DIGITAL DATA PROCESSING HANDLING RECORD CARRIERS PHYSICS PRESENTATION OF DATA RECOGNITION OF DATA RECORD CARRIERS |
| title | Text matching method based on data enhancement and graph matching network |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-12T01%3A33%3A20IST&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=LIU%20NING&rft.date=2022-12-23&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3ECN115510841A%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 |