{"id":378,"date":"2019-02-05T08:33:52","date_gmt":"2019-02-05T08:33:52","guid":{"rendered":"https:\/\/wordpress.cs.vt.edu\/cs4984spring19\/?p=378"},"modified":"2019-02-07T08:58:50","modified_gmt":"2019-02-07T08:58:50","slug":"reading-reflection-3-02-05-2019-liz-dao","status":"publish","type":"post","link":"https:\/\/wordpress.cs.vt.edu\/cs4984spring19\/2019\/02\/05\/reading-reflection-3-02-05-2019-liz-dao\/","title":{"rendered":"[Reading Reflection #3] \u2013 [02\/05\/2019] \u2013 [Liz Dao]"},"content":{"rendered":"\n

Early Public Responses to the\nZika-Virus on YouTube: Prevalence of and Differences Between Conspiracy Theory\nand Informational Videos<\/strong><\/p>\n\n\n\n

Summary:<\/strong><\/p>\n\n\n\n

As\nYouTube emerged as a popular social media platform, it also became the favorite\nplayground of fake news and conspiracy theory videos. Surprisingly, most of the\nvideos on this website are health-related. This creates a concern that fake\nnews and conspiracy theory content can mislead and install fear among the\naudience thus affecting the spread of epidemics. The researchers collected the\ncontent of the 35 most widespread YouTube videos during the first phase of the\nZika-virus outbreak \u2013 from December 30, 2015, to March 30, 2016 \u2013 and user\nreaction to those videos. This paper aims to find the dissimilarities in terms\nof user activity (number of comments, shares, likes, and dislikes), and the\nattitude and content of user responses between informational and conspiracy\nvideos. <\/p>\n\n\n\n

Unexpectedly,\nthere is no significant difference between these two types of videos in most\ncases. Both informational and conspiracy videos share not only the same amount\nof responses and unique users per view but also a low rate of additional\nresponding per unique user. Furthermore, the user comments on these two types\nof videos are slightly negative, which challenging Vousoughi, Roy and Aral\u2019s\nconclusion that false news provokes more negative sentiments than true news.\nHowever, there is a dissimilarity in the content of user response between\ninformational and conspiracy videos. Informational videos user responses focus\non the consequences of the virus. Whereas, the conspiracy theory videos user\nresponses are more interested in finding out who is responsible for the\noutbreak. <\/p>\n\n\n\n

Reflection:<\/strong><\/p>\n\n\n\n

    Even though the studying of misleading\nhealth-related content on YouTube might be worth researching, the authors\u2019\nchoice to limit their research to the topic Zika-virus is, sadly, not a wise\ndecision. First of all, despite the fact that YouTube has over one billion\nhighly active users watching around six billion hours of videos per month, it\nis also true that distribution of user age and interest are extremely skewed.\nAlso, a large percentage of YouTube users are teenagers and young adults who\nhave little or no interest in Zika-virus. In fact, there is a huge gap in user\nactivity and engagement between popular topics (gaming and beauty) and other\nvideos. Much of the popular health-related content (anorexia, tanning,\nvaccines, etc.) is also linking to beauty trends. In other words, YouTube is\nnot an ideal social media forum for studying user response to different type of\nvideos of Zika-virus or most health issues. The perfect proof for this is the\nextremely small dataset the authors were able to acquire on the topic. So, if the same question was asked about\nvideos related to vaccines, the result might have been more interesting. <\/strong><\/p>\n\n\n\n

    Another concern about the dataset is that\nthere is a lack of explanation and inspection of the data collected. What are the user statistics: age range,\ngender, prior behavior, etc.? Do the dataset includes videos that were reported\nas many conspiracy theory videos can be flagged as inappropriate? Why did the\nauthors choose 40,000 as the cut off for their dataset? How popular are\nZika-virus related videos compared to other videos posted on the same day?  <\/strong><\/p>\n\n\n\n

    As mentioned above, the most active group\nof YouTube users are not attracted to the topic of Zika-virus. Older users are\nless likely to show reactions on social media forums, especially on\ncontroversial topics. Therefore, the user inactivity and passiveness can make\nthe difference in user activity between informational and conspiracy theory\nvideos seem insignificant. Also, the low rate of additional responding can also\nbe the result of the user behavior on social media forums rather. There needs to be more analysis of the user\nbehavior before concluding that the user activity is similar across the two\ntypes of videos. <\/strong><\/p>\n\n\n\n

    The most interesting analysis of this\nresearch is the semantic maps of user comments between informational and\nconspiracy theory videos. The clusters of the informational videos are bigger\nand more concentrated. Surprisingly, offensive words (shit, ass, etc.) are used\nmore frequently in informational videos. Moreover, the comments in conspiracy\ntheory videos are more concerned with foreign forces such as Brazil, Africa,\netc. Meanwhile, the audience of informational videos focuses more on the\nnation\u2019s eternal conflicts between parties and religions. What might be the cause of the difference in interest and usage of offensive\nlanguage?<\/strong> The fact that Bill Gates, the only businessman, is not only\nmentioned but also has his name misspelled frequently is interesting. Why did\nhe appear as much as presidents of the United States in the comment? Does the common misspelling of his name\nindicate the education level of the audience?     <\/strong><\/p>\n\n\n\n

Automated Hate Speech Detection and\nthe Problem of Offensive Language<\/strong><\/p>\n\n\n\n

Summary:<\/strong><\/p>\n\n\n\n

    The paper address one of the biggest\nchallenges of automatic hate-speech detection: distinguish hate speech from\noffensive language. The authors define hate speech \u201cas a language that is used\nto expresses hatred towards a targeted group or is intended to be derogatory,\nto humiliate, or to insult the members of the group.\u201d Even with a stricter\ndefinition of hate speech than previous studies, the authors still find the\ndistinction between these two types of expression too ambiguous and case\nspecific. The raw data contains tweets with terms from the hate-speech lexicon,\nwhich is compiled by the website Hatebase.org. Then, the tweets were reviewed\nand grouped into three categories: hate-speech, offensive language, and none of\nthe above. After trials and errors, the authors decided to build a logistic\nregression model. Even though the algorithm achieves a relatively high overall\nprecision of 91%, it does not accurately differentiate hate-speech from\noffensive languages with a misclassification rate of 40%. <\/p>\n\n\n\n

    Despite the improvement in precision and\naccuracy compared to previous models, the research acknowledges that there are\nplenty of issues with the current model. The algorithm often flags tweets as\nless hateful or offensive than human coders. The heavy reliance on the presence\nof particular terms and neglect of the tone and context of the tweets result in\na high misclassification. Moreover, it is noticeable that sexist tweets are\nmuch less likely to be considered as hateful than racist or homophobic tweets.\nThe authors suggest future research should consider social context and\nconversations in which hate-speech arises. Also, studying the behavior and\nmotivation of hate speakers can also provide more insight into the\ncharacteristics of these expressions. On the other hand, another pitfall of the\ncurrent model is that it performs poorly on less common types of hate-speech\nsuch as those targeting Chinese immigrants. <\/p>\n\n\n\n

Reflection:<\/strong><\/p>\n\n\n\n

    It is not an exaggeration to say that even\nhumans are struggling to differentiate hateful speeches from those that are\nmerely offensive, let alone an algorithm. This research shows that the presence\nand frequency of keywords are by far sufficient in distinguishing hate speech.\nIndeed, in most cases, the decisive factors determining whether a tweet is\nhateful or offensive are the context of the conversation, the sentiment of the\ntweet, and the user speaking pattern. However, analyzing these factors is much\neasier said than done. <\/p>\n\n\n\n

    The authors suggest user attributes such as\ngender and ethnicity can be helpful in the classification of hate-speech. But\nacquiring these data might be impossible. Even in the few cases in which users\nagree to provide their demographic information, the trustworthiness is\nunguaranteed. However, these attributes can still be derived from analyzing the\nuser behavior in the forum. For example, a user who often likes and retweets\ncooking recipes and cat pictures have a higher chance to be a woman. Therefore, future studies can consider\nlooking into a user prior behavior to predict their tendency of expressing hate-speech\nand classify their tweets\u2019 nature. <\/strong><\/p>\n\n\n\n

    One of the main causes of the high\nmisclassification of hate-speech is that the word choices and writing styles vary\ngreatly among users. Even though most hateful expressions are aggressive and\ncontain many curse or offensive terms, there are plentiful exceptions. An\nexpression can still \u201cintended to be derogatory, to humiliate, or to insult\u201d\ndisadvantaged groups without using a single offensive term. In fact, this is\nthe most dangerous form of hate-speech as it often disguises itself as a\nstatement of truth or solid argument. On the other side, there are statements\nusing flagged terms yet convey positive messages. Therefore, rather than focusing on the vocabulary of the tweets, it\nmight be better to analyze the messages the users want to convey.  <\/strong><\/p>\n","protected":false},"excerpt":{"rendered":"

Early Public Responses to the Zika-Virus on YouTube: Prevalence of and Differences Between Conspiracy Theory and Informational Videos Summary: As YouTube emerged as a popular social media platform, it also became the favorite playground of fake news and conspiracy theory videos. Surprisingly, most of the videos on this website are health-related. This creates a concern […]<\/p>\n","protected":false},"author":246,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-378","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"jetpack_featured_media_url":"","_links":{"self":[{"href":"https:\/\/wordpress.cs.vt.edu\/cs4984spring19\/wp-json\/wp\/v2\/posts\/378","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/wordpress.cs.vt.edu\/cs4984spring19\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/wordpress.cs.vt.edu\/cs4984spring19\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/wordpress.cs.vt.edu\/cs4984spring19\/wp-json\/wp\/v2\/users\/246"}],"replies":[{"embeddable":true,"href":"https:\/\/wordpress.cs.vt.edu\/cs4984spring19\/wp-json\/wp\/v2\/comments?post=378"}],"version-history":[{"count":2,"href":"https:\/\/wordpress.cs.vt.edu\/cs4984spring19\/wp-json\/wp\/v2\/posts\/378\/revisions"}],"predecessor-version":[{"id":381,"href":"https:\/\/wordpress.cs.vt.edu\/cs4984spring19\/wp-json\/wp\/v2\/posts\/378\/revisions\/381"}],"wp:attachment":[{"href":"https:\/\/wordpress.cs.vt.edu\/cs4984spring19\/wp-json\/wp\/v2\/media?parent=378"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/wordpress.cs.vt.edu\/cs4984spring19\/wp-json\/wp\/v2\/categories?post=378"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/wordpress.cs.vt.edu\/cs4984spring19\/wp-json\/wp\/v2\/tags?post=378"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}