\u201cEmotion Classification with Visibility Graphs\u201d ba\u015fl\u0131kl\u0131 \u00e7al\u0131\u015fmam\u0131z IEEE Signal Processing Letters dergisinde yay\u0131mlanm\u0131\u015ft\u0131r.<\/p>\n
\u00c7al\u0131\u015fmada, d\u00f6n\u00fc\u015ft\u00fcr\u00fcc\u00fclerin sabit ba\u011flam pencerelerinin \u00f6tesine ge\u00e7en ve metin i\u00e7indeki genel ili\u015fkileri yakalayan, g\u00f6r\u00fcn\u00fcrl\u00fck \u00e7izgesine dayal\u0131 ard\u0131\u015f\u0131k bir modelleme yakla\u015f\u0131m\u0131 \u00f6nerilmektedir. Y\u00f6ntem, herhangi bir d\u00f6n\u00fc\u015ft\u00fcr\u00fcc\u00fc tabanl\u0131 modele eklenebilen mod\u00fcler bir \u00e7izge uzant\u0131s\u0131 olarak \u00e7al\u0131\u015fmakta ve bu modellerin SemEval2018 ile GoEmotions gibi yayg\u0131n b\u00fcy\u00fck \u00f6l\u00e7ekli veri setlerinde en g\u00fcncel y\u00f6ntemleri geride b\u0131rakmas\u0131n\u0131 sa\u011flamaktad\u0131r.<\/p>\n\n
Makale: https:\/\/ieeexplore.ieee.org\/document\/10985783<\/a> \u00d6zet:<\/strong><\/p>\n D\u00f6n\u00fc\u015ft\u00fcr\u00fcc\u00fcler, temsil g\u00fc\u00e7leri ve ba\u015far\u0131mlar\u0131 nedeniyle do\u011fal dil i\u015flemede \u00f6ne \u00e7\u0131km\u0131\u015ft\u0131r. D\u00f6n\u00fc\u015ft\u00fcr\u00fcc\u00fcler do\u011fal dili s\u0131n\u0131rl\u0131 ba\u011flam pencereleri \u00fczerinde bir dizi olarak i\u015fler, ancak bu pencerelerin \u00f6tesindeki s\u00f6zc\u00fckler aras\u0131 genel ili\u015fkiler yaln\u0131zca dizi i\u015fleme ile tam olarak modellenemez. Bu sorunu hafifletmek i\u00e7in, sinir a\u011flar\u0131na geometrik uzant\u0131lar getiren ve metin i\u00e7indeki ili\u015fkilerin \u00f6\u011frenilmesini sa\u011flayan \u00e7izge sinir a\u011f\u0131 (GNN) tabanl\u0131 modeller \u00f6nerilmi\u015ftir. Ancak s\u0131radan \u00e7izge tabanl\u0131 y\u00f6ntemler metnin ard\u0131\u015f\u0131k yap\u0131s\u0131n\u0131 g\u00f6z ard\u0131 eder. Bu \u00e7al\u0131\u015fmada, klasik zaman serisi bilgisini \u00e7izge temsillerine d\u00f6n\u00fc\u015ft\u00fcren g\u00f6r\u00fcn\u00fcrl\u00fck \u00e7izgelerini kullanan ilk genel \u00e7izge tabanl\u0131 sinir a\u011f\u0131 olan EmoVis \u00f6nerilmektedir. Problem bir duygu s\u0131n\u0131fland\u0131rma g\u00f6revi olarak ele al\u0131nm\u0131\u015f ve modelin etiketler ile c\u00fcmledeki s\u00f6zc\u00fckler aras\u0131ndaki ili\u015fkileri \u00f6\u011frenmesi sa\u011flanm\u0131\u015ft\u0131r. EmoVis, herhangi bir d\u00f6n\u00fc\u015ft\u00fcr\u00fcc\u00fc tabanl\u0131 modele mod\u00fcler bir \u00e7izge uzant\u0131s\u0131 olarak eklenebilmekte ve farkl\u0131 dillerde bu modellerin ba\u015far\u0131m\u0131n\u0131 ve \u00f6\u011frenme yetene\u011fini belirgin bi\u00e7imde art\u0131rmaktad\u0131r. Deneyler, EmoVis’in d\u00f6n\u00fc\u015ft\u00fcr\u00fcc\u00fc tabanl\u0131 modellerin SemEval2018 yar\u0131\u015fma veri setlerinde farkl\u0131 dillerdeki \u00fc\u00e7 veri setinde ve GoEmotions veri setinde en g\u00fcncel y\u00f6ntemleri geride b\u0131rakmas\u0131n\u0131 sa\u011flad\u0131\u011f\u0131n\u0131 g\u00f6stermektedir.<\/p>\n","protected":false},"excerpt":{"rendered":" \u201cEmotion Classification with Visibility Graphs\u201d ba\u015fl\u0131kl\u0131 \u00e7al\u0131\u015fmam\u0131z IEEE Signal Processing Letters dergisinde yay\u0131mlanm\u0131\u015ft\u0131r. \u00c7al\u0131\u015fmada, d\u00f6n\u00fc\u015ft\u00fcr\u00fcc\u00fclerin sabit ba\u011flam pencerelerinin \u00f6tesine ge\u00e7en ve metin i\u00e7indeki genel ili\u015fkileri yakalayan, g\u00f6r\u00fcn\u00fcrl\u00fck \u00e7izgesine dayal\u0131 ard\u0131\u015f\u0131k bir modelleme yakla\u015f\u0131m\u0131 \u00f6nerilmektedir. Y\u00f6ntem, herhangi bir d\u00f6n\u00fc\u015ft\u00fcr\u00fcc\u00fc tabanl\u0131 modele eklenebilen mod\u00fcler bir \u00e7izge uzant\u0131s\u0131 olarak \u00e7al\u0131\u015fmakta ve bu modellerin SemEval2018 ile GoEmotions gibi yayg\u0131n\u2026<\/p>\n","protected":false},"author":9,"featured_media":12718,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[445],"tags":[],"class_list":["post-13564","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-duyurular"],"_links":{"self":[{"href":"https:\/\/umram.bilkent.edu.tr\/index.php\/wp-json\/wp\/v2\/posts\/13564","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/umram.bilkent.edu.tr\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/umram.bilkent.edu.tr\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/umram.bilkent.edu.tr\/index.php\/wp-json\/wp\/v2\/users\/9"}],"replies":[{"embeddable":true,"href":"https:\/\/umram.bilkent.edu.tr\/index.php\/wp-json\/wp\/v2\/comments?post=13564"}],"version-history":[{"count":1,"href":"https:\/\/umram.bilkent.edu.tr\/index.php\/wp-json\/wp\/v2\/posts\/13564\/revisions"}],"predecessor-version":[{"id":13565,"href":"https:\/\/umram.bilkent.edu.tr\/index.php\/wp-json\/wp\/v2\/posts\/13564\/revisions\/13565"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/umram.bilkent.edu.tr\/index.php\/wp-json\/wp\/v2\/media\/12718"}],"wp:attachment":[{"href":"https:\/\/umram.bilkent.edu.tr\/index.php\/wp-json\/wp\/v2\/media?parent=13564"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/umram.bilkent.edu.tr\/index.php\/wp-json\/wp\/v2\/categories?post=13564"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/umram.bilkent.edu.tr\/index.php\/wp-json\/wp\/v2\/tags?post=13564"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
\nKod: https:\/\/github.com\/koc-lab\/EmoVis<\/a><\/p>\n\n