﻿{"id":6211,"date":"2018-12-27T22:43:19","date_gmt":"2018-12-27T14:43:19","guid":{"rendered":"http:\/\/www.nlpir.org\/wordpress\/?p=6211"},"modified":"2019-01-01T21:06:52","modified_gmt":"2019-01-01T13:06:52","slug":"3han-a-deep-neural-network-for-fake-news-detection","status":"publish","type":"post","link":"http:\/\/www.nlpir.org\/wordpress\/2018\/12\/27\/3han-a-deep-neural-network-for-fake-news-detection\/","title":{"rendered":"3HAN: A Deep Neural Network for Fake News Detection"},"content":{"rendered":"\n<h2 class=\"wp-block-heading\" style=\"text-align:center\">NLPIR SEMINAR Y2018#13<\/h2>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>INTRO<\/strong><\/h3>\n\n\n\n<p>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; In the new semester, our Lab, Web Search\nMining and Security Lab, plans to hold an academic seminar every Wednesdays,\nand each time a keynote speaker will share understanding of papers published in\nrecent years with you.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Arrangement<\/strong><\/h3>\n\n\n\n<p>This week&#8217;s seminar is organized as follows:<br>1. The seminar time is <strong>1.pm, Fri<\/strong>, at Zhongguancun Technology Park ,Building 5, 1306. <br>2. The lecturer is <strong>Ilham<\/strong>, the paper&#8217;s title is <strong>3HAN: A Deep Neural Network for Fake News Detection<\/strong>. <br>3. The seminar will be hosted by Zhaoyou Liu. <br>4. Attachment is the paper of this seminar, please download in advance. <br><br><\/p>\n\n\n\n<p>Anyone interested in this topic is welcomed to join us. the following is the abstract for this week\u2019s paper<\/p>\n\n\n\n<div style=\"border:dotted windowtext 1.0pt;padding:1.0pt 4.0pt 1.0pt 4.0pt;\">\n\t<p class=\"MsoNormal\" align=\"center\" style=\"text-align:center;\">\n\t\t<span style=\"font-size:18px;\">3HAN: A Deep Neural Network for Fake News Detection<\/span>\n\t<\/p>\n\t<p class=\"MsoNormal\" align=\"center\" style=\"text-align:center;\">\n\t\tSneha Singhania&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Nigel Fernandez&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; and Shrisha Rao\n\t<\/p>\n\t<p class=\"MsoNormal\" align=\"center\" style=\"text-align:center;\">\n\t\t<span style=\"font-size:16px;\">Abstract<\/span>\n\t<\/p>\n\t<p class=\"MsoNormal\">\n\t\t<span>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; The rapid spread of fake news is a\nserious problem calling for AI solutions. We employ a deep learning based\nautomated detector through a three level hierarchical attention network (3HAN)\nfor fast, accurate detection of fake news. 3HAN has three levels, one each for\nwords, sentences, and the headline, and constructs a news vector: an effective\nrepresentation of an input news article, by processing an article in an\nhierarchical bottom-up manner. The headline is known to be a distinguishing\nfeature of fake news, and furthermore, relatively few words and sentences in an\narticle are more important than the rest. 3HAN gives a differential importance\nto parts of an article, on account of its three layers of attention. By\nexperiments on a large real-world data set, we observe the effectiveness of\n3HAN with an accuracy of 96.77%. Unlike some other deep learning models, 3HAN\nprovides an understandable output through the attention weights given to\ndifferent parts of an article, which can be visualized through a heatmap to\nenable further manual fact checking.<\/span>\n\t<\/p>\n<\/div>\n\n\n\n<figure class=\"wp-block-image\"><img loading=\"lazy\" decoding=\"async\" width=\"659\" height=\"408\" src=\"http:\/\/www.nlpir.org\/wordpress\/wp-content\/uploads\/2018\/12\/20181227-\u4e0b\u5348-102152.jpg\" alt=\"\" class=\"wp-image-6222\" srcset=\"http:\/\/www.nlpir.org\/wordpress\/wp-content\/uploads\/2018\/12\/20181227-\u4e0b\u5348-102152.jpg 659w, http:\/\/www.nlpir.org\/wordpress\/wp-content\/uploads\/2018\/12\/20181227-\u4e0b\u5348-102152-300x186.jpg 300w, http:\/\/www.nlpir.org\/wordpress\/wp-content\/uploads\/2018\/12\/20181227-\u4e0b\u5348-102152-80x50.jpg 80w\" sizes=\"(max-width: 659px) 100vw, 659px\" \/><\/figure>\n\n\n\n<div class=\"wp-block-file\"><a href=\"http:\/\/www.nlpir.org\/wordpress\/wp-content\/uploads\/2018\/12\/3HAN-A-Deep-Neural-Network-for-Fake-News-Detection.pdf\">3HAN A Deep Neural Network for Fake News Detection<\/a><a href=\"http:\/\/www.nlpir.org\/wordpress\/wp-content\/uploads\/2018\/12\/3HAN-A-Deep-Neural-Network-for-Fake-News-Detection.pdf\" class=\"wp-block-file__button\" download>\u4e0b\u8f7d<\/a><\/div>\n\n\n\n<!--nextpage-->\n\n\n\n<h3 class=\"wp-block-heading\" style=\"text-align:center\"><strong>NLPIR\nSEMINAR 13th ISSUE COMPLETED<\/strong><\/h3>\n\n\n\n<p> \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0  Last week, Ilham gave a presentation about the paper, 3HAN: A Deep Neural Network for Fake News Detection, and shared some opinion on it.<\/p>\n\n\n\n<figure class=\"wp-block-image\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"768\" src=\"http:\/\/www.nlpir.org\/wordpress\/wp-content\/uploads\/2019\/01\/IMG_20181228_2-1024x768.jpg\" alt=\"\" class=\"wp-image-6286\" srcset=\"http:\/\/www.nlpir.org\/wordpress\/wp-content\/uploads\/2019\/01\/IMG_20181228_2-1024x768.jpg 1024w, http:\/\/www.nlpir.org\/wordpress\/wp-content\/uploads\/2019\/01\/IMG_20181228_2-300x225.jpg 300w, http:\/\/www.nlpir.org\/wordpress\/wp-content\/uploads\/2019\/01\/IMG_20181228_2-768x576.jpg 768w, http:\/\/www.nlpir.org\/wordpress\/wp-content\/uploads\/2019\/01\/IMG_20181228_2-200x150.jpg 200w, http:\/\/www.nlpir.org\/wordpress\/wp-content\/uploads\/2019\/01\/IMG_20181228_2-80x60.jpg 80w\" sizes=\"(max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p>After the presentation, several questions\nwere asked. The Q&amp;As are listed as following:<\/p>\n\n\n\n<p>Q:Is this paper dependent on content,  whether the title is more important? because there are many news&#8217; contents in China that do not match the title.<br>A: This paper uses 3HAN, it has three levels, one each for words, sentences, and the headline, and constructs a news vector, different parts of an article have a attention weights.<\/p>\n\n\n\n<p>Q: Does anyone have ideals about detecting fake news?<br>\nA: our jobs may not only be in detecting fake news, in social media , wo can find its original organization which organization posted this news by  its contents.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>NLPIR SEMINAR Y2018#13 INTRO &nbsp;&#038;nbsp &hellip; <a href=\"http:\/\/www.nlpir.org\/wordpress\/2018\/12\/27\/3han-a-deep-neural-network-for-fake-news-detection\/\">\u7ee7\u7eed\u9605\u8bfb <span class=\"meta-nav\">&rarr;<\/span><\/a><\/p>\n","protected":false},"author":862,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[37,38],"tags":[],"_links":{"self":[{"href":"http:\/\/www.nlpir.org\/wordpress\/wp-json\/wp\/v2\/posts\/6211"}],"collection":[{"href":"http:\/\/www.nlpir.org\/wordpress\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/www.nlpir.org\/wordpress\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/www.nlpir.org\/wordpress\/wp-json\/wp\/v2\/users\/862"}],"replies":[{"embeddable":true,"href":"http:\/\/www.nlpir.org\/wordpress\/wp-json\/wp\/v2\/comments?post=6211"}],"version-history":[{"count":6,"href":"http:\/\/www.nlpir.org\/wordpress\/wp-json\/wp\/v2\/posts\/6211\/revisions"}],"predecessor-version":[{"id":6289,"href":"http:\/\/www.nlpir.org\/wordpress\/wp-json\/wp\/v2\/posts\/6211\/revisions\/6289"}],"wp:attachment":[{"href":"http:\/\/www.nlpir.org\/wordpress\/wp-json\/wp\/v2\/media?parent=6211"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/www.nlpir.org\/wordpress\/wp-json\/wp\/v2\/categories?post=6211"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/www.nlpir.org\/wordpress\/wp-json\/wp\/v2\/tags?post=6211"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}