{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,2]],"date-time":"2026-04-02T13:57:38Z","timestamp":1775138258820,"version":"3.50.1"},"reference-count":98,"publisher":"Association for Computing Machinery (ACM)","issue":"CSCW1","license":[{"start":{"date-parts":[[2021,4,13]],"date-time":"2021-04-13T00:00:00Z","timestamp":1618272000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["IIS-1717330"],"award-info":[{"award-number":["IIS-1717330"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["Proc. ACM Hum.-Comput. Interact."],"published-print":{"date-parts":[[2021,4,13]]},"abstract":"<jats:p>Algorithmic timeline curation is now an integral part of Twitter's platform, affecting information exposure for more than 150 million daily active users. Despite its large-scale and high-stakes impact, especially during a public health emergency such as the COVID-19 pandemic, the exact effects of Twitter's curation algorithm generally remain unknown. In this work, we present a sock-puppet audit that aims to characterize the effects of algorithmic curation on source diversity and topic diversity in Twitter timelines. We created eight sock puppet accounts to emulate representative real-world users, selected through a large-scale network analysis. Then, for one month during early 2020, we collected the puppets' timelines twice per day. Broadly, our results show that algorithmic curation increases source diversity in terms of both Twitter accounts and external domains, even though it drastically decreases the number of external links in the timeline. In terms of topic diversity, algorithmic curation had a mixed effect, slightly amplifying a cluster of politically-focused tweets while squelching clusters of tweets focused on COVID-19 fatalities and health information. Finally, we present some evidence that the timeline algorithm may exacerbate partisan differences in exposure to different sources and topics. The paper concludes by discussing broader implications in the context of algorithmic gatekeeping.<\/jats:p>","DOI":"10.1145\/3449152","type":"journal-article","created":{"date-parts":[[2021,4,22]],"date-time":"2021-04-22T17:51:09Z","timestamp":1619113869000},"page":"1-28","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":46,"title":["More Accounts, Fewer Links"],"prefix":"10.1145","volume":"5","author":[{"given":"Jack","family":"Bandy","sequence":"first","affiliation":[{"name":"Northwestern University, Evanston, IL, USA"}]},{"given":"Nicholas","family":"Diakopoulos","sequence":"additional","affiliation":[{"name":"Northwestern University, Evanston, IL, USA"}]}],"member":"320","published-online":{"date-parts":[[2021,4,22]]},"reference":[{"key":"e_1_2_2_1_1","doi-asserted-by":"publisher","DOI":"10.1509\/jmr.08.0468"},{"key":"e_1_2_2_2_1","unstructured":"American Press Institute. 2020. What is the purpose of journalism? https:\/\/www.americanpressinstitute.org\/journalism-essentials\/what-is-journalism\/purpose-journalism\/"},{"key":"e_1_2_2_3_1","doi-asserted-by":"publisher","DOI":"10.1037\/0278--6133.22.1.60"},{"key":"e_1_2_2_4_1","doi-asserted-by":"publisher","DOI":"10.1145\/3308560.3316494"},{"key":"e_1_2_2_5_1","doi-asserted-by":"publisher","DOI":"10.1145\/3209581"},{"key":"e_1_2_2_6_1","volume-title":"The media monopoly","author":"Bagdikian Ben H","unstructured":"Ben H Bagdikian. 1983. The media monopoly. Boston."},{"key":"e_1_2_2_7_1","doi-asserted-by":"publisher","DOI":"10.1145\/1935826.1935845"},{"key":"e_1_2_2_8_1","doi-asserted-by":"publisher","DOI":"10.1126\/science.aaa1160"},{"key":"e_1_2_2_9_1","volume-title":"Auditing News Curation Systems: A Case Study Examining Algorithmic and Editorial Logic in Apple News. International conference on web and social media (ICWSM)","author":"Bandy Jack","year":"2020","unstructured":"Jack Bandy and Nicholas Diakopoulos. 2020. Auditing News Curation Systems: A Case Study Examining Algorithmic and Editorial Logic in Apple News. International conference on web and social media (ICWSM) (2020)."},{"key":"e_1_2_2_10_1","doi-asserted-by":"publisher","DOI":"10.1080\/21670811.2018.1510741"},{"key":"e_1_2_2_11_1","doi-asserted-by":"publisher","unstructured":"Marianne Bertrand and Sendhil Mullainathan. 2003. Are Emily and Greg More Employable than Lakisha and Jamal? A Field Experiment on Labor Market Discrimination. (2003). https:\/\/doi.org\/10.2139\/ssrn.422902","DOI":"10.2139\/ssrn.422902"},{"key":"e_1_2_2_12_1","doi-asserted-by":"publisher","DOI":"10.1088\/1742-5468"},{"key":"e_1_2_2_13_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10676-015-9380-y"},{"key":"e_1_2_2_14_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.chb.2010.02.008"},{"key":"e_1_2_2_15_1","doi-asserted-by":"publisher","DOI":"10.1080\/1369118X.2019.1637447"},{"key":"e_1_2_2_16_1","unstructured":"Axel Bruns. 2019 b. Are Filter Bubbles Real? John Wiley & Sons."},{"key":"e_1_2_2_17_1","doi-asserted-by":"publisher","DOI":"10.1093\/poq"},{"key":"e_1_2_2_18_1","volume-title":"Fabr\u00ed cio Benevenuto, and Krishna P Gummadi","author":"Cha Meeyoung","year":"2010","unstructured":"Meeyoung Cha, Hamed Haddadi, Fabr\u00ed cio Benevenuto, and Krishna P Gummadi. 2010. Measuring User Influence in Twitter: The Million Follower Fallacy. In ICWSM."},{"key":"e_1_2_2_19_1","unstructured":"Samantha Cole. 2020. How to Take Your Twitter Feed Back From the Algorithm . https:\/\/www.vice.com\/en_us\/article\/7kz9ez\/go-into-2020-by-taking-your-twitter-feed-back-from-the-algorithms"},{"key":"e_1_2_2_20_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.tele.2018.07.004"},{"key":"e_1_2_2_21_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10606-014-9208-z"},{"key":"e_1_2_2_22_1","doi-asserted-by":"publisher","DOI":"10.1145\/2872518.2889302"},{"key":"e_1_2_2_23_1","article-title":"Evaluating the Accuracy of Implicit Feedback from Clicks and Query Reformulations in Web Search","volume":"25","author":"Dept Thorsten Joachims","year":"2007","unstructured":"Thorsten Joachims Dept, Computer Science, Laura Granka Google Inc, Helene Hembrooke Dept, Information Science, Filip Radlinski Dept, and Geri G A Y Dept. 2007. Evaluating the Accuracy of Implicit Feedback from Clicks and Query Reformulations in Web Search. ACM Transactions on Information Systems, Vol. 25, 2 (2007). https:\/\/dl.acm.org\/doi\/abs\/10.1145\/1229179.1229181","journal-title":"ACM Transactions on Information Systems"},{"key":"e_1_2_2_24_1","doi-asserted-by":"publisher","DOI":"10.1080\/21670811.2016.1178592"},{"key":"e_1_2_2_25_1","doi-asserted-by":"publisher","DOI":"10.1145\/3025453.3025659"},{"key":"e_1_2_2_26_1","volume-title":"Automating the News: How Algorithms Are Rewriting the Media","author":"Diakopoulos Nicholas","unstructured":"Nicholas Diakopoulos. 2019. Automating the News: How Algorithms Are Rewriting the Media. Harvard University Press."},{"key":"e_1_2_2_27_1","unstructured":"Stuart Dredge. 2014. Yes Twitter is putting tweets in your timeline from people you don't follow. https:\/\/www.theguardian.com\/technology\/2014\/oct\/17\/twitter-tweets-timeline-dont-follow"},{"key":"e_1_2_2_28_1","doi-asserted-by":"publisher","DOI":"10.1111\/j.1460--2466.1993.tb01304.x"},{"key":"e_1_2_2_29_1","doi-asserted-by":"publisher","DOI":"10.1145\/2685553.2702690"},{"key":"e_1_2_2_30_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41562-020-00954-0"},{"key":"e_1_2_2_31_1","doi-asserted-by":"publisher","DOI":"10.1177\/1461444817724170"},{"key":"e_1_2_2_32_1","unstructured":"Agence France-Presse. 2020. Twitter shares surge on millions new users. https:\/\/tribune.net.ph\/index.php\/2020\/02\/07\/twitter-shares-surge-on-millions-new-users\/"},{"key":"e_1_2_2_34_1","doi-asserted-by":"publisher","DOI":"10.1037\/a0021981"},{"key":"e_1_2_2_35_1","doi-asserted-by":"publisher","DOI":"10.1145\/1008992.1009079"},{"key":"e_1_2_2_36_1","doi-asserted-by":"publisher","DOI":"10.1126\/science.aau2706"},{"key":"e_1_2_2_37_1","doi-asserted-by":"publisher","DOI":"10.1126\/science.aau2706"},{"key":"e_1_2_2_38_1","doi-asserted-by":"publisher","DOI":"10.1080\/21670811.2017.1338145"},{"key":"e_1_2_2_39_1","doi-asserted-by":"publisher","DOI":"10.1145\/2488388.2488435"},{"key":"e_1_2_2_40_1","doi-asserted-by":"publisher","DOI":"10.1080\/21670811.2019.1623700"},{"key":"e_1_2_2_41_1","doi-asserted-by":"publisher","DOI":"10.1080\/1369118X.2016.1271900"},{"key":"e_1_2_2_42_1","doi-asserted-by":"publisher","DOI":"10.1080\/17512781003640703"},{"key":"e_1_2_2_43_1","volume-title":"Less of the same: The lack of local news on the Internet. Prepared for the FCC","author":"Hindman Matthew","year":"2011","unstructured":"Matthew Hindman. 2011. Less of the same: The lack of local news on the Internet. Prepared for the FCC (2011)."},{"key":"e_1_2_2_44_1","doi-asserted-by":"publisher","DOI":"10.1145\/1964858.1964870"},{"key":"e_1_2_2_45_1","doi-asserted-by":"publisher","DOI":"10.1145\/3308558.3313654"},{"key":"e_1_2_2_46_1","volume-title":"Short Text Topic Modeling Techniques, Applications, and Performance: A Survey. (apr","author":"Jipeng Qiang","year":"2019","unstructured":"Qiang Jipeng, Qian Zhenyu, Li Yun, Yuan Yunhao, and Wu Xindong. 2019. Short Text Topic Modeling Techniques, Applications, and Performance: A Survey. (apr 2019). arxiv: 1904.07695 http:\/\/arxiv.org\/abs\/1904.07695"},{"key":"e_1_2_2_47_1","unstructured":"Mark Jurkowitz and Amy Mitchell. 2020. A fifth of Democrats Republicans get news only from outlets with like-minded audiences. https:\/\/www.journalism.org\/2020\/03\/04\/about-one-fifth-of-democrats-and-republicans-get-political-news-in-a-kind-of-media-bubble\/"},{"key":"e_1_2_2_49_1","doi-asserted-by":"publisher","DOI":"10.1145\/2926720"},{"key":"e_1_2_2_50_1","unstructured":"Jacob Kastrenakes. 2020. Twitter says AI tweet recommendations helped it add millions of users. https:\/\/www.theverge.com\/2020\/2\/6\/21125431\/twitter-q4--2019-earnings-daily-user-growth-machine-learning"},{"key":"e_1_2_2_51_1","unstructured":"Nicolas Koumchatzky and Anton Andryeyev. 2017. Using Deep Learning at Scale in Twitter's Timelines. https:\/\/blog.twitter.com\/engineering\/en_us\/topics\/insights\/2017\/using-deep-learning-at-scale-in-twitters-timelines.html"},{"key":"e_1_2_2_52_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10791-018-9341-2"},{"key":"e_1_2_2_53_1","doi-asserted-by":"publisher","DOI":"10.1145\/1772690.1772751"},{"key":"e_1_2_2_54_1","doi-asserted-by":"publisher","unstructured":"David Lazer. 2020. Studying human attention on the Internet. 21--22 pages. https:\/\/doi.org\/10.1073\/pnas.1919348117","DOI":"10.1073\/pnas.1919348117"},{"key":"e_1_2_2_55_1","doi-asserted-by":"publisher","DOI":"10.1145\/3308558.3313682"},{"key":"e_1_2_2_56_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.pubrev.2017.10.002"},{"key":"e_1_2_2_57_1","unstructured":"Nicholas L\u00e9 onard and Cibele Montez Halasz. 2018. Twitter meets TensorFlow. https:\/\/blog.twitter.com\/engineering\/en_us\/topics\/insights\/2018\/twittertensorflow.html"},{"key":"e_1_2_2_58_1","doi-asserted-by":"publisher","DOI":"10.1145\/2911451.2911499"},{"key":"e_1_2_2_59_1","doi-asserted-by":"publisher","DOI":"10.1145\/3340631.3394864"},{"key":"e_1_2_2_60_1","unstructured":"Ingrid Lunden. 2020. Twitter Q1: sales up 3% to $808M as it swings to a loss on COVID-19 mDAUS hit record 166M. https:\/\/techcrunch.com\/2020\/04\/30\/twitter-q1-sales-up-3-to-808m-as-it-swigs-to-a-loss-on-covid-19-mdaus-hit-record-166m\/"},{"key":"e_1_2_2_61_1","doi-asserted-by":"publisher","DOI":"10.1080\/21670811.2015.1133249"},{"key":"e_1_2_2_63_1","doi-asserted-by":"publisher","DOI":"10.1145\/2484028.2484166"},{"key":"e_1_2_2_64_1","doi-asserted-by":"publisher","DOI":"10.1145\/3359231"},{"key":"e_1_2_2_66_1","volume-title":"Proceedings of the 9th International Conference on Web and Social Media, ICWSM","author":"Mitra Tanushree","year":"2015","unstructured":"Tanushree Mitra and Eric Gilbert. 2015. CREDBANK: A large-scale social media corpus with associated credibility annotations. In Proceedings of the 9th International Conference on Web and Social Media, ICWSM 2015. 258--267. https:\/\/www.aaai.org\/ocs\/index.php\/ICWSM\/ICWSM15\/paper\/viewPaper\/10582"},{"key":"e_1_2_2_67_1","doi-asserted-by":"publisher","DOI":"10.1080\/1369118X.2018.1444076"},{"key":"e_1_2_2_68_1","doi-asserted-by":"publisher","DOI":"10.5325\/jinfopoli.1.2011.0246"},{"key":"e_1_2_2_69_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.telpol.2014.12.003"},{"key":"e_1_2_2_70_1","volume-title":"Social Media and the Public Interest: The Rise of Algorithmic News and the Future of the Marketplace of Ideas. Telecommunications Policy","author":"Napoli Philip M.","year":"2015","unstructured":"Philip M. Napoli. 2015b. Social Media and the Public Interest: The Rise of Algorithmic News and the Future of the Marketplace of Ideas. Telecommunications Policy (2015)."},{"key":"e_1_2_2_71_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.chb.2018.07.043"},{"key":"e_1_2_2_72_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cosrev.2018.07.001"},{"key":"e_1_2_2_73_1","doi-asserted-by":"publisher","DOI":"10.5860\/CHOICE.50-0926"},{"key":"e_1_2_2_74_1","volume-title":"Fifth international AAAI conference on weblogs and social media (ICWSM).","author":"Pennacchiotti Marco","year":"2011","unstructured":"Marco Pennacchiotti and Ana-Maria Popescu. 2011. A Machine Learning Approach to Twitter User Classification. In Fifth international AAAI conference on weblogs and social media (ICWSM)."},{"key":"e_1_2_2_75_1","doi-asserted-by":"publisher","DOI":"10.17645\/mac.v8i3.3001"},{"key":"e_1_2_2_76_1","doi-asserted-by":"publisher","DOI":"10.1145\/2441955.2441981"},{"key":"e_1_2_2_77_1","doi-asserted-by":"publisher","DOI":"10.1145\/3274417"},{"key":"e_1_2_2_78_1","doi-asserted-by":"publisher","DOI":"10.1177\/0002764203259293"},{"key":"e_1_2_2_79_1","volume-title":"Auditing Algorithms: Research Methods for Detecting Discrimination on Internet Platforms. In Data and discrimination: converting critical concerns into productive inquiry. 1--23.","author":"Sandvig Christian","year":"2014","unstructured":"Christian Sandvig, Kevin Hamilton, Karrie Karahalios, and Cedric Langbort. 2014. Auditing Algorithms: Research Methods for Detecting Discrimination on Internet Platforms. In Data and discrimination: converting critical concerns into productive inquiry. 1--23."},{"key":"e_1_2_2_80_1","volume-title":"The Power of News","unstructured":"Michael. Schudson. 1995. The Power of News. Harvard University Press. 269 pages."},{"key":"e_1_2_2_81_1","doi-asserted-by":"publisher","DOI":"10.1111\/jcc4.12013"},{"key":"e_1_2_2_82_1","volume-title":"How the Twitter Algorithm Works","author":"Social Sprout","year":"2020","unstructured":"Sprout Social. 2020. How the Twitter Algorithm Works in 2020. https:\/\/sproutsocial.com\/insights\/twitter-algorithm\/"},{"key":"e_1_2_2_83_1","volume-title":"Examining the Alternative Media Ecosystem through the Production of Alternative Narratives of Mass Shooting Events on Twitter. ICWSM","author":"Starbird Kate","year":"2017","unstructured":"Kate Starbird. 2017. Examining the Alternative Media Ecosystem through the Production of Alternative Narratives of Mass Shooting Events on Twitter. ICWSM (2017)."},{"key":"e_1_2_2_84_1","doi-asserted-by":"publisher","DOI":"10.1145\/3134920"},{"key":"e_1_2_2_85_1","doi-asserted-by":"publisher","DOI":"10.1111\/j.1460--2466.2010.01497.x"},{"key":"e_1_2_2_86_1","doi-asserted-by":"publisher","DOI":"10.2196\/jmir.6045"},{"key":"e_1_2_2_87_1","doi-asserted-by":"publisher","DOI":"10.1145\/3290605.3300683"},{"key":"e_1_2_2_88_1","doi-asserted-by":"publisher","DOI":"10.1080\/1369118X.2020.1764605"},{"key":"e_1_2_2_89_1","unstructured":"Twitter. 2020 a. About your Twitter timeline. https:\/\/help.twitter.com\/en\/using-twitter\/twitter-timeline"},{"key":"e_1_2_2_90_1","unstructured":"Twitter. 2020 b. Twitter moments: Guidelines and principles. https:\/\/help.twitter.com\/en\/rules-and-policies\/twitter-moments-guidelines-and-principles https:\/\/about.twitter.com\/company\/moments-guidelines"},{"key":"e_1_2_2_91_1","doi-asserted-by":"publisher","DOI":"10.1080\/08838151.2017.1344673"},{"key":"e_1_2_2_92_1","volume-title":"Proceedings of the International AAAI Conference on Web and Social Media (ICWSM).","author":"Vincent Nicholas","year":"2019","unstructured":"Nicholas Vincent, Isaac Johnson, Patrick Sheehan, and Brent Hecht. 2019. Measuring the Importance of User-Generated Content to Search Engines. In Proceedings of the International AAAI Conference on Web and Social Media (ICWSM)."},{"key":"e_1_2_2_93_1","doi-asserted-by":"publisher","DOI":"10.1126\/science.aap9559"},{"key":"e_1_2_2_94_1","doi-asserted-by":"publisher","DOI":"10.1080\/17512786.2016.1171162"},{"key":"e_1_2_2_95_1","doi-asserted-by":"publisher","unstructured":"James G Webster and Harsh Taneja. 2018. Building and interpreting audience networks: A response to mukerjee majo-vazquez & gonzalez-bailon. E11--E14 pages. https:\/\/doi.org\/10.1093\/joc\/jqy024","DOI":"10.1093\/joc"},{"key":"e_1_2_2_96_1","doi-asserted-by":"publisher","DOI":"10.1177\/1461444818807133"},{"key":"e_1_2_2_97_1","volume-title":"Sizing Up Twitter Users","author":"Wojcik Stefan","year":"2019","unstructured":"Stefan Wojcik and Adam Hughes. 2019. Sizing Up Twitter Users. Pew Research Center (2019), 1--23."},{"key":"e_1_2_2_98_1","doi-asserted-by":"publisher","DOI":"10.1145\/1963405.1963504"},{"key":"e_1_2_2_99_1","doi-asserted-by":"publisher","DOI":"10.1177\/1464884918791349"},{"key":"e_1_2_2_100_1","volume-title":"Scalable and Generalizable Social Bot Detection through Data Selection","author":"Yang Kai-Cheng","unstructured":"Kai-Cheng Yang, Onur Varol, Pik-Mai Hui, and Filippo Menczer. 2020. Scalable and Generalizable Social Bot Detection through Data Selection. In Association for the Advancement of Artificial Intelligence (AAAI)."},{"key":"e_1_2_2_101_1","doi-asserted-by":"publisher","DOI":"10.1145\/2623330.2623715"},{"key":"e_1_2_2_102_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-20161-5_34"}],"container-title":["Proceedings of the ACM on Human-Computer Interaction"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3449152","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3449152","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3449152","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T20:18:17Z","timestamp":1750191497000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3449152"}},"subtitle":["How Algorithmic Curation Impacts Media Exposure in Twitter Timelines"],"short-title":[],"issued":{"date-parts":[[2021,4,13]]},"references-count":98,"journal-issue":{"issue":"CSCW1","published-print":{"date-parts":[[2021,4,13]]}},"alternative-id":["10.1145\/3449152"],"URL":"https:\/\/doi.org\/10.1145\/3449152","relation":{},"ISSN":["2573-0142"],"issn-type":[{"value":"2573-0142","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,4,13]]},"assertion":[{"value":"2021-04-22","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}