{"id":391,"date":"2020-11-14T21:00:01","date_gmt":"2020-11-14T21:00:01","guid":{"rendered":"https:\/\/www.electionanalysis.ws\/us\/?page_id=391"},"modified":"2020-11-15T09:29:28","modified_gmt":"2020-11-15T09:29:28","slug":"detecting-emotions-in-facebook-political-ads-with-computer-vision","status":"publish","type":"page","link":"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/","title":{"rendered":"Detecting emotions in Facebook political ads with computer vision"},"content":{"rendered":"\n<div class=\"wp-block-columns is-layout-flex wp-container-core-columns-is-layout-9d6595d7 wp-block-columns-is-layout-flex\">\n<div class=\"wp-block-column is-layout-flow wp-block-column-is-layout-flow\" style=\"flex-basis:25%\">\n<figure class=\"wp-block-image size-large\"><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" width=\"179\" height=\"179\" src=\"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_headshot.png?resize=179%2C179&#038;ssl=1\" alt=\"\" class=\"wp-image-392\" srcset=\"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_headshot.png?w=179&amp;ssl=1 179w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_headshot.png?resize=150%2C150&amp;ssl=1 150w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_headshot.png?resize=60%2C60&amp;ssl=1 60w\" sizes=\"auto, (max-width: 179px) 100vw, 179px\" \/><\/figure>\n\n\n\n<p class=\"US205bio\"><strong>Dr Michael Bossetta<\/strong><br><br>Assistant Professor in the Department of Communication and Media at Lund University. His research interests revolve around the intersection of social media and politics, especially political campaigning. He produces and hosts the Social Media and Politics podcast, available for free on any podcast app.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" width=\"453\" height=\"453\" src=\"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_co-author_Schm_kel_headshot.png?resize=453%2C453&#038;ssl=1\" alt=\"\" class=\"wp-image-393\" srcset=\"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_co-author_Schm_kel_headshot.png?w=453&amp;ssl=1 453w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_co-author_Schm_kel_headshot.png?resize=300%2C300&amp;ssl=1 300w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_co-author_Schm_kel_headshot.png?resize=150%2C150&amp;ssl=1 150w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_co-author_Schm_kel_headshot.png?resize=60%2C60&amp;ssl=1 60w\" sizes=\"auto, (max-width: 453px) 100vw, 453px\" \/><\/figure>\n\n\n\n<p class=\"US205bio\"><strong>Rasmus Schm\u00f8kel<br><\/strong><br>Copenhagen University M.Sc. in Political Science from Copenhagen University. His research interests include visual political communication, computational methods, and the role of emotions in algorithmic bias.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" width=\"400\" height=\"308\" src=\"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/US20_divider_5_social_media.png?resize=400%2C308&#038;ssl=1\" alt=\"\" class=\"wp-image-190\" srcset=\"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/US20_divider_5_social_media.png?w=400&amp;ssl=1 400w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/US20_divider_5_social_media.png?resize=300%2C231&amp;ssl=1 300w\" sizes=\"auto, (max-width: 400px) 100vw, 400px\" \/><\/figure>\n\n\n<h5 class=\"US205\"> Section 5: Social media<\/h5>\n<div class=\"page-list\"><ul class=\"list-group-item\"><li class=\"page_item page-item-328\"><a href=\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/media-and-social-media-platforms-finally-begin-to-embrace-their-roles-as-democratic-gatekeepers\/\">Media and social media platforms finally begin to embrace their roles as democratic gatekeepers<\/a><\/li>\n<li class=\"page_item page-item-331\"><a href=\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/did-social-media-make-us-more-or-less-politically-unequal-in-2020\/\">Did social media make us more or less politically unequal in 2020?<\/a><\/li>\n<li class=\"page_item page-item-335\"><a href=\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/platform-transparency-in-the-fight-against-disinformation\/\">Platform transparency in the fight against disinformation<\/a><\/li>\n<li class=\"page_item page-item-341\"><a href=\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/why-trumps-determination-to-sow-doubt-about-data-undermines-democracy\/\">Why Trump\u2019s determination to sow doubt about data undermines democracy<\/a><\/li>\n<li class=\"page_item page-item-344\"><a href=\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/a-banner-year-for-advertising-and-a-look-at-differences-across-platforms\/\">A banner year for advertising and a look at differences across platforms<\/a><\/li>\n<li class=\"page_item page-item-350\"><a href=\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/how-joe-biden-conveyed-empathy\/\">How Joe Biden conveyed empathy<\/a><\/li>\n<li class=\"page_item page-item-353\"><a href=\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/the-debates-and-the-election-conversation-on-twitter\/\">The debates and the election conversation on Twitter<\/a><\/li>\n<li class=\"page_item page-item-356\"><a href=\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/did-the-economy-covid-19-or-black-lives-matter-to-the-senate-candidates-in-2020\/\">Did the economy, COVID-19, or Black Lives Matter to the Senate candidates in 2020?<\/a><\/li>\n<li class=\"page_item page-item-362\"><a href=\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/leadership-through-showmanship-trumps-ability-to-coin-nicknames-for-opponents-on-twitter\/\">Leadership through showmanship: Trump\u2019s ability to coin nicknames for opponents on Twitter<\/a><\/li>\n<li class=\"page_item page-item-366\"><a href=\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/election-countdown-instagrams-role-in-visualizing-the-2020-campaign\/\">Election countdown: Instagram\u2019s role in visualizing the 2020 campaign<\/a><\/li>\n<li class=\"page_item page-item-371\"><a href=\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/candidates-did-lackluster-youth-targeting-on-instagram\/\">Candidates did lackluster youth targeting on Instagram<\/a><\/li>\n<li class=\"page_item page-item-374\"><a href=\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/college-students-political-engagement-and-snapchat-in-the-2020-general-election\/\">College students, political engagement and Snapchat in the 2020 general election<\/a><\/li>\n<li class=\"page_item page-item-381\"><a href=\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/advertising-on-facebook-transparency-but-not-transparent-enough\/\">Advertising on Facebook: transparency, but not transparent enough<\/a><\/li>\n<\/ul><\/div>\n<\/div>\n\n\n\n<div class=\"wp-block-column is-layout-flow wp-block-column-is-layout-flow\" style=\"flex-basis:75%\">\n<p>Set against the backdrop of a pandemic and polarized political climate, this election was sure to be emotional. Therefore, we used computer vision to examine the emotions expressed by Biden and Trump in their Facebook political ads. Our aim was to see if the candidates expressed different emotions in images where they presented themselves versus how they depicted opponents in attacks.<\/p>\n\n\n\n<p>Indeed, we found that Biden primarily expressed happiness in his Facebook ads, whereas his attack ads depicted Trump as angry. Meanwhile, the Trump campaign presented their candidate as a calm leader, while Biden was often shown expressing confusion. Overall, both campaigns\u2019 Facebook ads largely focused more on promoting their own candidate rather than on attacking the opponent.<\/p>\n\n\n\n<p>To collect and analyze images in Facebook ads, we used two open science software tools that we developed. The first,&nbsp;<a href=\"https:\/\/github.com\/schmokel\/FBAdLibrarian\">FBAdLibrarian<\/a>, assists researchers in collecting images from the Facebook Ad Library. The second,&nbsp;<a href=\"https:\/\/github.com\/schmokel\/pykognition\">Pykognition<\/a>, leverages Amazon\u2019s facial detection algorithms to classify emotions expressed in faces.<\/p>\n\n\n\n<p>We collected ads from the official Trump and Biden Facebook pages in the week before Facebook\u2019s ad pause on October 27th. Due to controversies around this ad pause, the data we collected from Facebook may be incomplete, and it appears that Facebook removed all information about dates from the data.<\/p>\n\n\n\n<p>In total, we obtained 202,000 Trump ads and 109,287 Biden ads. Using FBAdLibrarian, we collected 98,830 images from Trump ads and 69,941 images from Biden ads. This means that in our dataset, 49% of Trump\u2019s ads and 64% of Biden\u2019s ads contained still images (the remainder were videos).<\/p>\n\n\n\n<p>We then removed all images that did not depict Trump or Biden. This includes ads that were infographics, promoting merchandise, or featuring other high-profile politicians. Many of the remaining images were copies, so we removed duplicates and ended up with a dataset of 634 images for Trump and 1,148 for Biden.<\/p>\n\n\n\n<p>We further divided these images into three ad categories used in&nbsp;<a href=\"https:\/\/www.cambridge.org\/core\/journals\/american-political-science-review\/article\/political-advertising-online-and-offline\/9E24E81AC74E4644494FF451D5373B71\">previous research<\/a>: Promote, Contrast, and Attack. Promote images show only the candidate, Contrast images show both the candidate and the opponent, and Attack images show only the opponent. We divided the images this way to see whether campaigns strategically change the emotions displayed by their candidate versus the opponent.<\/p>\n\n\n\n<p><strong>Emotion Classification Results<\/strong><\/p>\n\n\n\n<p>We ran all images through Amazon\u2019s Rekognition API with our software, Pykognition. This process categorizes faces into eight emotional categories: Angry, Calm, Confused, Disgusted, Fear, Happy, Sad, and Surprised. Each face is classified with a unique identifier (&#8220;FaceID&#8221;), an emotion, and a predicted score for that emotion. In Figure 1, we show examples from the Biden page for each category: Promote, Contrast, and Attack.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><a href=\"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/BOSSETTA_FIgure_1.png?ssl=1\"><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" width=\"648\" height=\"208\" src=\"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/BOSSETTA_FIgure_1.png?resize=648%2C208&#038;ssl=1\" alt=\"\" class=\"wp-image-394\" srcset=\"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/BOSSETTA_FIgure_1.png?resize=1024%2C329&amp;ssl=1 1024w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/BOSSETTA_FIgure_1.png?resize=300%2C96&amp;ssl=1 300w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/BOSSETTA_FIgure_1.png?resize=768%2C247&amp;ssl=1 768w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/BOSSETTA_FIgure_1.png?resize=1536%2C494&amp;ssl=1 1536w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/BOSSETTA_FIgure_1.png?w=1885&amp;ssl=1 1885w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/BOSSETTA_FIgure_1.png?w=1296&amp;ssl=1 1296w\" sizes=\"auto, (max-width: 648px) 100vw, 648px\" \/><\/a><figcaption><strong>Figure 1: Promote, Contrast, and Attack ad examples from the Biden Facebook page<\/strong><\/figcaption><\/figure>\n\n\n\n<p>We manually checked the algorithm\u2019s classification for each image. In cases where we disagreed with the algorithm, we changed the emotion to the one we considered most accurate. Overall, we agreed with the algorithm in 73% of cases.<\/p>\n\n\n\n<p>Interestingly and unique to this election cycle, candidates (and Biden in particular) wore protective facemasks in response to the coronavirus pandemic. This proved problematic for the facial detection algorithm, which predicted pictures with facemasks to display emotions such as sadness and fear but with unreliable confidence. We therefore reclassified all images where candidates wore a mask into a new category: &#8220;MASK&#8221;. With the images classified, we were able to link our coded data to all other ads in the dataset using the same image. In Figure 2, we present our overall results per candidate page and for each ad type.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><a href=\"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_Figure_2_Corrected.png?ssl=1\"><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" width=\"648\" height=\"261\" src=\"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_Figure_2_Corrected.png?resize=648%2C261&#038;ssl=1\" alt=\"\" class=\"wp-image-395\" srcset=\"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_Figure_2_Corrected.png?resize=1024%2C412&amp;ssl=1 1024w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_Figure_2_Corrected.png?resize=300%2C121&amp;ssl=1 300w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_Figure_2_Corrected.png?resize=768%2C309&amp;ssl=1 768w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_Figure_2_Corrected.png?resize=1536%2C619&amp;ssl=1 1536w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_Figure_2_Corrected.png?w=1840&amp;ssl=1 1840w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_Figure_2_Corrected.png?w=1296&amp;ssl=1 1296w\" sizes=\"auto, (max-width: 648px) 100vw, 648px\" \/><\/a><figcaption><strong><strong>Figure 2: Overall emotion classifications by candidate and ad type<\/strong><\/strong><\/figcaption><\/figure>\n\n\n\n<p>Overall, we find that in terms of the number of ads sent, both campaigns emphasized promoting their own candidate rather than attacking the opponent. Biden was most often depicted as \u2018Happy\u2019, and he wore a facemask in approximately 10% of images. By contrast, the Trump page issued ads promoting the candidate as the \u2018Calm\u2019 leader and wore a mask in less than 1% of images. In Figure 3, we also report each category in proportional terms, to better understand the distribution of emotions per ad type.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><a href=\"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_FIgure_3_Corrected.png?ssl=1\"><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" width=\"648\" height=\"256\" src=\"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_FIgure_3_Corrected.png?resize=648%2C256&#038;ssl=1\" alt=\"\" class=\"wp-image-396\" srcset=\"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_FIgure_3_Corrected.png?resize=1024%2C405&amp;ssl=1 1024w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_FIgure_3_Corrected.png?resize=300%2C119&amp;ssl=1 300w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_FIgure_3_Corrected.png?resize=768%2C304&amp;ssl=1 768w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_FIgure_3_Corrected.png?resize=1536%2C608&amp;ssl=1 1536w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_FIgure_3_Corrected.png?w=1865&amp;ssl=1 1865w, https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_FIgure_3_Corrected.png?w=1296&amp;ssl=1 1296w\" sizes=\"auto, (max-width: 648px) 100vw, 648px\" \/><\/a><figcaption><strong>Figure 3: Proportion of emotion classifications by candidate and ad type<\/strong><\/figcaption><\/figure>\n\n\n\n<p>For Biden, there was a clear distinction between his own portrayal as &#8220;Happy&#8221; and Trump as &#8220;Angry&#8221;. For Trump, we see his page\u2019s ads depicting him as &#8220;Calm&#8221;, whereas Biden was most often depicted as &#8220;Confused&#8221; in attacks.<\/p>\n\n\n\n<p>Our analysis reveals how campaigns preferred to show their candidates in Facebook ads: Biden as warm and happy, and Trump as the calm leader. In addition, we see how the campaign\u2019s broader attack narratives were also depicted in images: Biden attacked Trump as an angry despot, and Trump attacked Biden as a confused candidate in mental decline.<\/p>\n\n\n\n<p>It is important to note that we did not factor in the spending amount or number of people who saw these ads; we only studied the raw number of ads issued by the campaign. And, due to limitations imposed by Facebook, we do not know the dates of when these ads were issued. Nevertheless, we encourage other researchers to use the open science tools that we developed to further analyze emotions in political images.<\/p>\n<\/div>\n<\/div>\n","protected":false},"excerpt":{"rendered":"<p>Dr Michael Bossetta Assistant Professor in the Department of Communication and Media at Lund University. His research interests revolve around the intersection of social media and politics, especially political campaigning. He produces and hosts the Social Media and Politics podcast, available for free on any podcast app. Rasmus Schm\u00f8kelCopenhagen University M.Sc. in Political Science from [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"parent":22,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-391","page","type-page","status-publish","hentry"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v26.3 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>Detecting emotions in Facebook political ads with computer vision - Election Analysis - United States<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Detecting emotions in Facebook political ads with computer vision - Election Analysis - United States\" \/>\n<meta property=\"og:description\" content=\"Dr Michael Bossetta Assistant Professor in the Department of Communication and Media at Lund University. His research interests revolve around the intersection of social media and politics, especially political campaigning. He produces and hosts the Social Media and Politics podcast, available for free on any podcast app. Rasmus Schm\u00f8kelCopenhagen University M.Sc. in Political Science from [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/\" \/>\n<meta property=\"og:site_name\" content=\"Election Analysis - United States\" \/>\n<meta property=\"article:modified_time\" content=\"2020-11-15T09:29:28+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_headshot.png\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data1\" content=\"5 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/\",\"url\":\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/\",\"name\":\"Detecting emotions in Facebook political ads with computer vision - Election Analysis - United States\",\"isPartOf\":{\"@id\":\"https:\/\/www.electionanalysis.ws\/us\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_headshot.png\",\"datePublished\":\"2020-11-14T21:00:01+00:00\",\"dateModified\":\"2020-11-15T09:29:28+00:00\",\"breadcrumb\":{\"@id\":\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/#primaryimage\",\"url\":\"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_headshot.png?fit=179%2C179&ssl=1\",\"contentUrl\":\"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_headshot.png?fit=179%2C179&ssl=1\",\"width\":179,\"height\":179},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/www.electionanalysis.ws\/us\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"U.S. Election Analysis 2020\",\"item\":\"https:\/\/www.electionanalysis.ws\/us\/president2020\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"Section 5: Social media\",\"item\":\"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/\"},{\"@type\":\"ListItem\",\"position\":4,\"name\":\"Detecting emotions in Facebook political ads with computer vision\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.electionanalysis.ws\/us\/#website\",\"url\":\"https:\/\/www.electionanalysis.ws\/us\/\",\"name\":\"Election Analysis - United States\",\"description\":\"Election Analysis reports from the U.S. Presidential Elections\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.electionanalysis.ws\/us\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Detecting emotions in Facebook political ads with computer vision - Election Analysis - United States","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/","og_locale":"en_US","og_type":"article","og_title":"Detecting emotions in Facebook political ads with computer vision - Election Analysis - United States","og_description":"Dr Michael Bossetta Assistant Professor in the Department of Communication and Media at Lund University. His research interests revolve around the intersection of social media and politics, especially political campaigning. He produces and hosts the Social Media and Politics podcast, available for free on any podcast app. Rasmus Schm\u00f8kelCopenhagen University M.Sc. in Political Science from [&hellip;]","og_url":"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/","og_site_name":"Election Analysis - United States","article_modified_time":"2020-11-15T09:29:28+00:00","og_image":[{"url":"https:\/\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_headshot.png","type":"","width":"","height":""}],"twitter_card":"summary_large_image","twitter_misc":{"Est. reading time":"5 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/","url":"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/","name":"Detecting emotions in Facebook political ads with computer vision - Election Analysis - United States","isPartOf":{"@id":"https:\/\/www.electionanalysis.ws\/us\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/#primaryimage"},"image":{"@id":"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/#primaryimage"},"thumbnailUrl":"https:\/\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_headshot.png","datePublished":"2020-11-14T21:00:01+00:00","dateModified":"2020-11-15T09:29:28+00:00","breadcrumb":{"@id":"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/#primaryimage","url":"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_headshot.png?fit=179%2C179&ssl=1","contentUrl":"https:\/\/i0.wp.com\/www.electionanalysis.ws\/us\/wp-content\/uploads\/sites\/2\/2020\/11\/Michael_Bossetta_-_BOSSETTA_headshot.png?fit=179%2C179&ssl=1","width":179,"height":179},{"@type":"BreadcrumbList","@id":"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/detecting-emotions-in-facebook-political-ads-with-computer-vision\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.electionanalysis.ws\/us\/"},{"@type":"ListItem","position":2,"name":"U.S. Election Analysis 2020","item":"https:\/\/www.electionanalysis.ws\/us\/president2020\/"},{"@type":"ListItem","position":3,"name":"Section 5: Social media","item":"https:\/\/www.electionanalysis.ws\/us\/president2020\/section-5-social-media\/"},{"@type":"ListItem","position":4,"name":"Detecting emotions in Facebook political ads with computer vision"}]},{"@type":"WebSite","@id":"https:\/\/www.electionanalysis.ws\/us\/#website","url":"https:\/\/www.electionanalysis.ws\/us\/","name":"Election Analysis - United States","description":"Election Analysis reports from the U.S. Presidential Elections","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.electionanalysis.ws\/us\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"}]}},"jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/www.electionanalysis.ws\/us\/wp-json\/wp\/v2\/pages\/391","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.electionanalysis.ws\/us\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/www.electionanalysis.ws\/us\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/www.electionanalysis.ws\/us\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.electionanalysis.ws\/us\/wp-json\/wp\/v2\/comments?post=391"}],"version-history":[{"count":3,"href":"https:\/\/www.electionanalysis.ws\/us\/wp-json\/wp\/v2\/pages\/391\/revisions"}],"predecessor-version":[{"id":523,"href":"https:\/\/www.electionanalysis.ws\/us\/wp-json\/wp\/v2\/pages\/391\/revisions\/523"}],"up":[{"embeddable":true,"href":"https:\/\/www.electionanalysis.ws\/us\/wp-json\/wp\/v2\/pages\/22"}],"wp:attachment":[{"href":"https:\/\/www.electionanalysis.ws\/us\/wp-json\/wp\/v2\/media?parent=391"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}