<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Dániel Molnár on Medium]]></title>
        <description><![CDATA[Stories by Dániel Molnár on Medium]]></description>
        <link>https://medium.com/@molnardan95?source=rss-11309e4eb1e4------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*M4CF9p-pfoiE7vzqp5NIZg.jpeg</url>
            <title>Stories by Dániel Molnár on Medium</title>
            <link>https://medium.com/@molnardan95?source=rss-11309e4eb1e4------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 17:14:05 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@molnardan95/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[How biased are NFL teams’ websites in their coverage?]]></title>
            <link>https://medium.com/@molnardan95/how-biased-are-nfl-teams-websites-in-their-coverage-d12728e2fe33?source=rss-11309e4eb1e4------2</link>
            <guid isPermaLink="false">https://medium.com/p/d12728e2fe33</guid>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[sports]]></category>
            <category><![CDATA[data-analysis]]></category>
            <dc:creator><![CDATA[Dániel Molnár]]></dc:creator>
            <pubDate>Sat, 17 Nov 2018 08:14:21 GMT</pubDate>
            <atom:updated>2018-11-17T08:14:21.302Z</atom:updated>
            <content:encoded><![CDATA[<h3>How biased are NFL teams’ websites in their coverage? — Analysis using R and the aws.comprehend package</h3><p><em>Disclaimer: This article has been written as part of a university homework. Furher analysis is needed to get meaningful results.</em></p><p>AWS Comprehend is a cloud based natural language processing service, that can perform different text analysis tasks, including sentiment analysis. To use this package, there are two main things needed. First of all, you will need an AWS account so you can utilize AWS to analyse your text. There you will need to set up a keypair for yourself, using the IAM Mangement Console — make sure to save the keypair, because AWS uses this to identify users accessing its services from R. Secondly, you will need R itself, where you will need to enter your credentials, so AWS can recognize you and you can use its services.</p><p>The main steps you will need to take is to install the required packages (aws.comprehend and rvest for data scaping, if you want to analyse a websites content like I did), and set up your environment with your credentials, using the <em>Sys.setenv()</em> function.</p><blockquote>Sys.setenv(“AWS_ACCESS_KEY_ID” = AWS_ACCESS_KEY_ID,<br> “AWS_SECRET_ACCESS_KEY” = AWS_SECRET_ACCESS_KEY,<br> “AWS_DEFAULT_REGION” = “eu-west-2”)</blockquote><p>From this point, you will utilize the aws.comprehend package to perform text analysis on whatever you want. For sentiment analysis, the <em>detect_sentiment()</em> function will be used, where the only mandatory parameter is the text itself, you would like to analyse. The are several supported languages, one of them being English, but you can find more information about the details of this package on CRAN.</p><h4><strong>Analysing NFL teams sites</strong></h4><p>Using the above mentioned method, I tried analysing the coverage of the two opposing teams of an NFL match. The example I used was the Cowboys’ 27–20 victory over the Eagles on 11/11, 2018 (mainly because I’m a ‘Boys fan).</p><p>I scaped the two articles which served as a game recap on the two teams websites. Then, I tried to analyse the sentiment of each paragraph one by one. Finally, I visualized the results in a scatter plot. <em>Blue dots are for the Cowboys, green one for the Eagles.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mjs99bKSNdCvs0HOR8vDEQ.png" /></figure><p>As we can see, there is some difference between the sentiment of the two articles, but not that much. If anything, we can tell that this was a very close match between the two teams, and the Cowboys probably won it — however, we can’t really be certain just from this one graph.</p><p>Analysing the sentiment even more, we see minor differences between the two articles. According to our results, the Cowboys paragraphs are positive with 10.8% certanity, whereas the Eagles ones with only 8.9%. The most important takeaway to answer our research question however is that both these articles seem pretty neutral. When checking the neutrality variable of these two game recaps, the ‘Boys one is 78% neutral, and the Eagles on is over 80%.</p><p>Alltogether, judging by these results, we could say that the websites’ coverage is unbiased as these articles seem to be very neutral. However, because we only analysed two articles and those for a match that was really close, external validity is questionable. Additional research is needed to tell if site neutrality is achieved and to what degree.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d12728e2fe33" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>