<?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 Adi Sidapara on Medium]]></title>
        <description><![CDATA[Stories by Adi Sidapara on Medium]]></description>
        <link>https://medium.com/@adisidapara?source=rss-98e294cc65a8------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*BFP14qZHmj_F9TWOGsO_Bg.jpeg</url>
            <title>Stories by Adi Sidapara on Medium</title>
            <link>https://medium.com/@adisidapara?source=rss-98e294cc65a8------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 04:02:22 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@adisidapara/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[Our Platform]]></title>
            <link>https://medium.com/@adisidapara/our-platform-238cfe1d775a?source=rss-98e294cc65a8------2</link>
            <guid isPermaLink="false">https://medium.com/p/238cfe1d775a</guid>
            <category><![CDATA[stanford]]></category>
            <category><![CDATA[college]]></category>
            <dc:creator><![CDATA[Adi Sidapara]]></dc:creator>
            <pubDate>Mon, 01 Apr 2019 16:18:38 GMT</pubDate>
            <atom:updated>2019-04-01T16:18:38.856Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Z_M4WBIbhEqvbJkeop5uew.jpeg" /></figure><p>Dear <strong>Class of 2022</strong>,</p><p>Now that we are two quarters into freshman year, we’ve had a lot of time to reflect. We’ve met some really awesome people, made amazing memories, taken classes that have pushed our boundaries, and have started to find in some capacity the communities here that we call home. But we’ve also come to realize how fragile that community can be. Life on this campus moves fast — friendships are constantly changing, and balance is hard to maintain. At times, it’s difficult to find your community when you’re still trying to figure out who you are. And those challenges will only grow during sophomore year. Life will change fundamentally for our grade: we won’t have our freshman year dorms, and all of us will disperse to very different pockets across campus. But that doesn’t mean that we should lose the support that we got from being freshmen.</p><p>We believe that community is a commitment — it’s something to be cultivated, something to be upheld, something to which we constantly strive. We may not be able to fully prevent the “sophomore slump,” but we can and will work to make that transition easier for everyone. We wanted to create a platform with that commitment in mind:</p><ol><li><strong>Strengthening and improving freshman communities. </strong>We would plan freshman dorm reunions (for West and East campus), de-stressing events (e.g. Field days, winter quarter rainy day slip-n-slides, petting zoos, free merch giveaways — our list of ideas goes on).</li><li><strong>Making all class events free.</strong> We want to make sure there are absolutely no barriers to students attending class events, assessing peer input to plan events that actually appeal to our class.</li><li><strong>Immersion in the community beyond the Bubble: </strong>We will forge partnerships with food places and other vendors off-campus to give the class discounts or promotions.</li><li><strong>Demystifying Stanford: </strong>We will be keeping a website where we post updates (upcoming events, meeting minutes, general tips, etc.) and get input back from you all.</li></ol><ul><li><strong>Consolidating information through a weekly newsletter.</strong> No one reads extraneous emails. We want to start a newspaper that’s catered toward our class in terms of relevant events and resources on campus.</li><li><strong>An information guide.</strong> We want to debunk myths, demystify mental health and Title IX resources, and offer support for our class in navigating life at Stanford.</li></ul><p>Ultimately, we want everyone to have a voice in shaping their sophomore experience. This coming week, during the active campaign season, we will be traveling to different dorms to understand what you all want the most from us as your class presidents.</p><p>Thank you for taking the time to consider our platform, and we are so excited by the prospect of serving you all. Here’s to an amazing end of freshman year and an even better sophomore year!</p><p>Sincerely,</p><p><strong>The Dream Team</strong></p><p>Adi Sidapara, Tiffany Liu, Mira Guleri, and Karsten Householder</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=238cfe1d775a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Thoughts on Standardizing Content Moderation (Part 2)]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/moonshot-blog/thoughts-on-standardizing-content-moderation-part-2-9d0515977398?source=rss-98e294cc65a8------2"><img src="https://cdn-images-1.medium.com/max/1090/1*kkWHK_G7YYAaIaZfq6BBSA.png" width="1090"></a></p><p class="medium-feed-snippet">To pick up where we left off, we last talked about the challenges that Facebook faces when it comes to moderating content. If you&#x2019;re&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/moonshot-blog/thoughts-on-standardizing-content-moderation-part-2-9d0515977398?source=rss-98e294cc65a8------2">Continue reading on Moonshots Blog »</a></p></div>]]></description>
            <link>https://medium.com/moonshot-blog/thoughts-on-standardizing-content-moderation-part-2-9d0515977398?source=rss-98e294cc65a8------2</link>
            <guid isPermaLink="false">https://medium.com/p/9d0515977398</guid>
            <category><![CDATA[social-media]]></category>
            <category><![CDATA[content-moderation]]></category>
            <category><![CDATA[tech-policy]]></category>
            <category><![CDATA[tech-regulation]]></category>
            <dc:creator><![CDATA[Adi Sidapara]]></dc:creator>
            <pubDate>Mon, 25 Mar 2019 23:01:25 GMT</pubDate>
            <atom:updated>2019-03-26T02:30:27.336Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[A Review of Facebook’s Content Moderation Strategy (Part 1)]]></title>
            <link>https://medium.com/moonshot-blog/a-review-of-facebooks-content-moderation-strategy-part-1-e2425f1d46ae?source=rss-98e294cc65a8------2</link>
            <guid isPermaLink="false">https://medium.com/p/e2425f1d46ae</guid>
            <category><![CDATA[user-experience]]></category>
            <category><![CDATA[content-moderation]]></category>
            <category><![CDATA[facebook]]></category>
            <dc:creator><![CDATA[Adi Sidapara]]></dc:creator>
            <pubDate>Tue, 29 Jan 2019 17:54:11 GMT</pubDate>
            <atom:updated>2019-01-29T17:55:32.885Z</atom:updated>
            <content:encoded><![CDATA[<h3>A Review of Facebook’s Content Moderation (Part 1)</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BuwmuQpLRsRjRo6BVVNkwA.png" /></figure><p>The weaponizing of fake news on social media platforms has been a wake up call to many of the big tech giants and regulatory authorities of our society. While the term itself was popularized by President Trump, its reality today is rather stark in many different contexts of media and politics.</p><p>Fake news has been around for quite a while. Since Egyptian and Roman times, political leaders such as Rameses and Octavian have <a href="https://www.ft.com/content/aaf2bb08-dca2-11e6-86ac-f253db7791c6">used misinformation</a> and <a href="https://www.publishersweekly.com/978-1-59233-336-3">misleading propaganda</a> to maintain political power and shame opponents and dissenters. Fake news continued through medieval and modern times into the 19th century United States, where yellow journalism written by Joseph Pulitzer and William Hearst aided the general public in prodding the U.S. government towards the Spanish-American war. By the 20th century, anti-German and anti-Nazi propaganda was used to pushing the public to be in favor of World War I and II and encouraging people to join the war effort in any way they could.</p><p>This past weekend, Facebook announced a proposal for its <a href="https://fbnewsroomus.files.wordpress.com/2019/01/draft-charter-oversight-board-for-content-decisions-1.pdf">new independent content regulation board</a>, one that would consist of around 40 “experts”. This proposal is a followup to <a href="https://qz.com/1465898/mark-zuckerberg-facebook-to-have-a-supreme-court-within-a-year/">the rumors</a> that Facebook was setting up its own internal “Supreme Court” to handle high-profile content disputes, and a follow up to its larger fact-checking program deployed in the United Kingdom in partnership with Full Fact, an independent fact checking organization.</p><p>The fact checking program is part of an expanding initiative by Facebook to limit the number of news articles and pieces containing inaccurate information, mainly as a response to criticism of Facebook’s role in spreading fake news during the 2016 presidential election. To check for “fake news,” Facebook partners with third-party fact checking institutions in each country that are certified by the International Fact-Checking Network. In the United States, for example, these <a href="https://www.facebook.com/help/publisher/182222309230722">organizations</a> include Associated Press, Factcheck.org, PolitiFact, and Snopes.com. After identifying fake news and reviewing their content, Facebook lowers its position in News Feed in order to ensure that less people see it and restrict/ban offenders who repeatedly post stories with misinformation.</p><p>Facebook’s history with fake news has not always been smooth. In March 2017, Facebook <a href="http://nymag.com/intelligencer/2017/12/facebook-gets-rid-of-disputed-flag-system-for-fake-news.html">implemented</a> a “disputed flag” system, allowing its own panel of fact-checkers to “flag” dodgy or fishy news stories in order to warn users about the credibility of some of the content on their news feed. However, it did nothing to stop the spread of “fake news”; in fact, the traffic around some “fake news” stories actually increased due to the attention it received by some people who argued that Facebook was trying to silence some news sources. Facebook got rid of the system quickly and replaced it with News Feed context, which allows users to get more information about a news source/author by clicking a small info button next to a news article; they say that they’re are implementing this system in order to help users evaluate the credibility of a source by themselves.</p><p>So how will Facebook’s fake news initiative have impact on government and public policy? Undoubtedly large social media sites such as Facebook and Twitter have the capability to influence a wide demographic of users, as was demonstrated in the 2016 presidential election with Russian interference. And this topic does not seem to be going away any time soon; <a href="https://www.nytimes.com/2018/04/10/us/politics/mark-zuckerberg-testimony.html">Mark Zuckerberg was called to testify</a> in front of lawmakers on Capitol Hill early last year and was questioned on Facebook’s role influencing political elections and the privacy of its users, and also faced tough questions on whether Facebook should be allowed to self-regulate or have restrictions placed by the U.S. government. In a <a href="https://www.wsj.com/articles/the-facts-about-facebook-11548374613">recent Op-Ed</a> published in the <em>Wall Street Journal</em>, he outlined the exact ways Facebook was utilizing their users’ data, stating that Facebook gives “people complete control over whether [they] use their information for ads, but [they] don’t let them control how [they] use it for security or operating their services.” He goes on to reiterate that Facebook has strong incentive to remove clickbait and divisive content and systems currently in place are still not perfect, yet doesn’t go into detail about how they plan on improving their systems. As of now, there has been no action by Congress to mandate or regulate fact-checking on social media sites like Facebook, and we’ll just have to trust that Facebook’s chosen third-party fact-checkers.</p><p>The dynamic between the way Facebook operates internally and its role in our society is one to marvel. While having independent boards is a first-step for big tech companies to subside to some sort of regulation on their platforms, it is disappointing to see the lack of user authority on a platform seeded on user community. If only there was a way to give users more authority in self-regulation, we would possibly see an increase in the trust they have for the platform (something Facebook needs, especially with the negative news cycles it has endured this past year) and less reliance on internal content moderation teams (which is <a href="https://motherboard.vice.com/en_us/article/xwk9zd/how-facebook-content-moderation-works">apparently close</a> to 7,500 people, and probably growing after increased scrutiny).</p><p>To be continued…</p><p><em>About the Authors</em></p><p><em>Charles Pan and Adi Sidapara are students at Stanford University and co-editors of Moonshots. Charles is studying Mathematical and Computational Sciences, and Adi is studying Computer Science and Public Policy.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e2425f1d46ae" width="1" height="1" alt=""><hr><p><a href="https://medium.com/moonshot-blog/a-review-of-facebooks-content-moderation-strategy-part-1-e2425f1d46ae">A Review of Facebook’s Content Moderation Strategy (Part 1)</a> was originally published in <a href="https://medium.com/moonshot-blog">Moonshots Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What is Moonshots?]]></title>
            <link>https://medium.com/moonshot-blog/what-is-moonshots-925aa38936fb?source=rss-98e294cc65a8------2</link>
            <guid isPermaLink="false">https://medium.com/p/925aa38936fb</guid>
            <category><![CDATA[government-innovation]]></category>
            <category><![CDATA[tech]]></category>
            <category><![CDATA[introduction]]></category>
            <dc:creator><![CDATA[Adi Sidapara]]></dc:creator>
            <pubDate>Tue, 01 Jan 2019 21:01:58 GMT</pubDate>
            <atom:updated>2019-01-08T06:05:01.326Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*R1DZdOR02PF4TcDBHy4q8w.png" /></figure><p>An engineering marvel, an unprecedented government-industry collaborative, a futuristic uplift : sending a man to the moon was the highlight of a generation. Apollo 11 was the original moonshot, but it inspired millions of people to take charge and led even more of its kind.</p><p>Moonshots is an experimental blog that covers the interplay between government and technology. While there are many independent writers that cover this growing intersection, student voices are often overlooked. Our belief is that combining the perspectives of students and industry experts can help create more engaging and relatable content for our target audience of students and young professionals.</p><p>The way our government interacts with technology today has been riddled with complacency and sluggishness. It is important that future leaders in technology and government realize the importance of the social implications in innovation and its relevant policymaking.</p><p>We plan on posting thought-provoking ideas on technology and government once a week. We want to include opinion pieces, fact-finding exposés, data journalism, and even summarized academic reviews. We are also open to exploring new types of intriguing and inspiring content along the way.</p><p>If our content theme excites you, we would love to have you contribute! We plan on announcing a call for submissions from college students every 10–12 weeks, but until our experimental cycle ends, we will keep it open for anyone at anytime. You can send us pitches by emailing moonshoteditors@gmail.com.</p><p>Sincerely,</p><p><a href="https://twitter.com/adisidapara">Adi Sidapara</a> and <a href="https://www.linkedin.com/in/charleslpan/">Charles Pan</a> (Founding Editors)</p><p>Edit: We originally intended to post on a strict timeline of every Monday, but as college students, our schedules are tight. We agreed to have a soft commitment for posting once a week. This frequency will change as time progresses with the blog, and we better realize the commitment of running this blog.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=925aa38936fb" width="1" height="1" alt=""><hr><p><a href="https://medium.com/moonshot-blog/what-is-moonshots-925aa38936fb">What is Moonshots?</a> was originally published in <a href="https://medium.com/moonshot-blog">Moonshots Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Bandak Lul: A Journey Through It All]]></title>
            <link>https://medium.com/@adisidapara/bandak-lul-a-journey-through-it-all-a58cd7cc2b5a?source=rss-98e294cc65a8------2</link>
            <guid isPermaLink="false">https://medium.com/p/a58cd7cc2b5a</guid>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[refugee-camps]]></category>
            <category><![CDATA[refugees]]></category>
            <category><![CDATA[immigrants]]></category>
            <category><![CDATA[refugee-crisis]]></category>
            <dc:creator><![CDATA[Adi Sidapara]]></dc:creator>
            <pubDate>Sun, 15 Oct 2017 16:47:16 GMT</pubDate>
            <atom:updated>2017-10-15T16:54:40.001Z</atom:updated>
            <content:encoded><![CDATA[<p>The American dream has been embodied by hundreds of millions of immigrants — yet the glorified image and narrative is quite different from the tumultuous adversity faced by many during their transitions. We are excited to announce the Immigrant in Action series for Unresolved, where we hope to share the inspiring stories of refugees and immigrants in communities across the world.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yIVDRfKxuzy8wUP9aKoTXg.jpeg" /><figcaption>Bandak Lul (Photo Credits to ASU New College)</figcaption></figure><p>Today, we are introducing Bandak Lul (Dak). Dak was born in South Sudan, but at the early age of 11 months, his family was forced to seek asylum in Ethiopia. During his time in Ethiopia, he stayed in a refugee camp — plagued by inadequate access to education and rationed food supplies, his formative years were untraditional; yet regardless, he is still grateful for his host country and the UNHCR for their humanitarian assistance.</p><p>At the age of 15, Dak, his sister, and his sister’s husband were granted asylum protection in the United States. They settled in Phoenix, AZ, where he attended secondary school at Camelback High School. As a new high school student, he found it difficult to integrate with the existing curriculum; the language barriers alone truly restricted him from achieving a quality education. However, with sheer dedication, he was determined — Dak enrolled in ESL classes during school and with a loaned computer from the school, learned how to type. Nevertheless, Dak graduated Camelback with a 4.0 GPA; he was offered a full ride scholarship by his school district to Phoenix College where he got his associates degree in political science. He furthered his education by transferring to Arizona State University (ASU) and earning a bachelor’s degree in political science and sociology. He also earned a Master’s degree in Social Justice &amp; Human Rights this summer from ASU — focused on refugee and human trafficking studies. ASU continues making Dak feels at home. Dak is now working for Arizona State University Office of Sex Trafficking Intervention Research (STIR) as a research project manager.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-6xcPrFRT0C1L4LlnfUCFQ.jpeg" /><figcaption>Bandak Graduating with his Master’s</figcaption></figure><p>Dak is more than an intellectual — he is a social and a human rights activist at large. He is part of the Lutheran Immigration and Refugee Services Leadership Academy with whom he lobbies Congress to hear refugee stories and voices. While working in the “big picture” issues in the refugee communities through political activism, he regularly volunteers in the local refugee community of Phoenix, connecting with them through their common identity of fleeing their homes due to the fear of persecution.</p><p>For Dak, technology has changed the way he communicates with back home. During his time in Ethiopia, he used to handwrite letters and send them through the unreliable postal service. When he moved to the United States, he worked through minimum wage jobs to earn and save enough money to buy a phone, with which he could communicate with his family. Through learning how to type and understand the English language, computers have aided Dak in transitioning to the United States. Dak is hopeful that through connecting refugees to technology and learning, we can aid and assist many with the adversity of finding a new home.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a58cd7cc2b5a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[More than Just Words: Unfolding Presidential Rhetoric]]></title>
            <link>https://medium.com/@adisidapara/more-than-just-words-unfolding-presidential-rhetoric-16edb29d1c74?source=rss-98e294cc65a8------2</link>
            <guid isPermaLink="false">https://medium.com/p/16edb29d1c74</guid>
            <category><![CDATA[big-data]]></category>
            <category><![CDATA[presidential-debates]]></category>
            <category><![CDATA[nlp]]></category>
            <category><![CDATA[sentiment-analysis]]></category>
            <category><![CDATA[2016-election]]></category>
            <dc:creator><![CDATA[Adi Sidapara]]></dc:creator>
            <pubDate>Thu, 13 Oct 2016 17:32:55 GMT</pubDate>
            <atom:updated>2016-10-14T08:27:34.517Z</atom:updated>
            <content:encoded><![CDATA[<p>Natural Language Processing Sentiment Analysis of Rhetoric from the Second Presidential Debate</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/880/1*LIwUUFiYYnqi4fpQt5n4JA.jpeg" /><figcaption>Credits to <a href="http://www.hellertoon.com/main.html">Joe Heller</a></figcaption></figure><h4><strong>Preface</strong></h4><p>American politics has come to a crossroads of embittering polarization and sardonic hypocrisy. With the ease-of-access to technology, our presidential candidates’ Twitter-itchy fingers carefreely post crude and lewd tweets and rock the foundations of political courtesy and decency that much of the world once looked upon as model election behavior. This fierce election which is (unfortunately) closely contested has seen both sides utter irrevocable rhetoric, with Trump referring to Mexicans as “<strong>rapists</strong>” and “<strong>criminals</strong>” and Hillary referring to Trump’s supporters as “<strong>irredeemable deplorables</strong>”. Certainly, no side is right — this culture of blatant invective has trickled down to everyday Americans. And with increased discourse on the community-sensitive topic of race relations, this election’s harsh words has only deepened the divisions among ourselves at a time when they need to be mended. This is why it is of the utmost importance to evaluate this election’s rhetoric, as it already has set precedent for future generations.</p><h4>Introduction</h4><p>In an increasingly data-driven world, machine learning has risen as a primary method of understanding and utilizing data for human good. A subfield of machine learning is natural language processing (NLP), and <a href="http://blog.algorithmia.com/introduction-natural-language-processing-nlp/">Algorithmia</a> defines it as “a way for computers to analyze, understand, and derive meaning from human language in a smart and useful way.” By parsing sentences for their atomic elements, NLP allows us to performing numerous tasks, one being sentiment analysis, computationally categorizing opinions and determining if they are positive, negative, or neutral to a topic.</p><p>Moreover, American politics for long has been branded as “negative politics”. Gallup found that most voters are discontent with negative campaigning. To determine to what extent this year’s election was negative, I performed naive statistical sentiment analysis on public campaign rhetoric of both candidates to compare and determine which candidate engaged the most in creating a negative political scene. I looked at the 2nd presidential debate as the foundation for my evaluation because it involved both candidates and engaged a larger audience then most individual campaign events.</p><h4>Methods</h4><p>In order to data-mine, I built <a href="https://www.dropbox.com/sh/7awu5778prwr05x/AADqNx_6ayU3vZDgf5PWoE7-a?dl=0">datasets</a> for both Clinton and Trump based on their respective portions in a <a href="http://fortune.com/2016/10/09/presidential-debate-read-transcript-donald-trump-hillary-clinton/">publicly available transcript</a> of the debate made available by Fortune. These were tokenized into sentences, and one interesting observation was that Trump had <em>significantly</em> more sentences than Hillary, exactly <em>1.9x</em> more.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OSC3RZb7JQ0FDH4Ed9tRsQ.png" /></figure><p>This largely reflects on Trump’s rather short sentence construction (which was incomplete often times).</p><p>The sentiment analysis library I used is <a href="https://cran.r-project.org/web/packages/syuzhet/index.html">Syuzhet</a>, a package for R. According to the author, Matthew Jockers, the library parses sentences to determine “emotional shifts” in sentiment rather than shifts in topic (a popular method). The package itself uses 4 sentiment dictionaries, which it refers for keywords and their respective scores.</p><blockquote>The Syuzhet package attempts to reveal the latent structure of narrative by means of sentiment analysis. Instead of detecting shifts in the topic or subject matter of the narrative (<a href="http://sappingattention.blogspot.com/2014/12/fundamental-plot-arcs-seen-through.html">as Ben Schmidt has done</a>), the Syuzhet package reveals the emotional shifts that serve as proxies for the narrative movement between conflict and conflict resolution.</blockquote><p>This is consequently used to generate a score, with the sign determining positive/negative sentiment and the number signifying a “magnitude”.</p><pre>&gt; library(syuzhet)<br>&gt; test.string &lt;- &quot;I love pie.&quot;<br>&gt; // package method that parses large strings for sentences and <br>&gt; //returns as a character vector<br>&gt; tokenized &lt;- get_sentences(test.string)<br>&gt; test.sent &lt;- get_sentiment(tokenized)<br>&gt; test.sent</pre><p>The test code above resulted in the following sentiment value.</p><pre>[1] 0.75</pre><p>For a negative sentence construct, the method returns a negative value.</p><pre>&gt; library(syuzhet)<br>&gt; test.string &lt;- &quot;I hate pie.&quot;<br>&gt; // package method that parses large strings for sentences and <br>&gt; //returns as a character vector<br>&gt; tokenized &lt;- get_sentences(test.string)<br>&gt; test.sent &lt;- get_sentiment(tokenized)<br>&gt; test.sent<br>[1] -0.75</pre><p>And to test the method with complex sentences, I used one more test.</p><pre>&gt; library(syuzhet)<br>&gt; test.string &lt;- &quot;I love myself like Kanye loves Kanye&quot;<br>&gt; // package method that parses large strings for sentences and <br>&gt; //returns as a character vector<br>&gt; tokenized &lt;- get_sentences(test.string)<br>&gt; test.sent &lt;- get_sentiment(tokenized)<br>&gt; test.sent<br>[1] 2.25</pre><p>Seems legit, right? Well, no. The shortcomings of the method arise when evaluating double negations.</p><pre>&gt; library(syuzhet)<br>&gt; test.string &lt;- &quot;I do not hate pie.&quot;<br>&gt; // package method that parses large strings for sentences and <br>&gt; //returns as a character vector<br>&gt; tokenized &lt;- get_sentences(test.string)<br>&gt; test.sent &lt;- get_sentiment(tokenized)<br>&gt; test.sent<br>[1] -0.75</pre><p>This hurdle did not affect the results too much as people rarely use double negations.</p><h4>Results</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/1*l7tFv9gKFvxKfh0rdwR_Kg.png" /><figcaption>Density Distribution of Hillary Clinton’s Sentiment Scores</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/1*ZAydx4mFnZluVZgA5YAKYA.png" /><figcaption>Density Distribution of Donald Trump’s Sentiment Scores</figcaption></figure><p>Assuming that anything below 0 is negative, anything above is positive, and anything at 0 is neutral, the sentiment score distribution for Trump and Clinton was interesting. Donald Trump had a median at 0, but after accounting for his outliers on the extreme negative, <strong>nearly 63% of his sentences scored negative</strong>. Hillary Clinton had a median at 0.25, and accounting for her negative outliers, <strong>around 47% of her sentences scored negative</strong>. This means that <strong>Trump was <em>1.34x more negative</em></strong> in the debate solely based on rhetoric.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PHOslLJwFJGcyIl4dvs0Ig.png" /><figcaption>Box Plot of Sentiment Score Distribution</figcaption></figure><p>The box plot offers a comparative display of the sentiment score distribution for each of the candidates. Trump had 26 outliers and Clinton had 16 outliers, indicating that Trump’s rhetoric fluctuated to both extremes more than Clinton’s. Out of these outliers, 9 were extreme positives and 7 were extreme negatives for Clinton, whereas for Trump, 12 were extreme positives and 14 were extreme negatives. In general, Trump’s more extreme sentiment rhetoric tended to be negative whereas Clinton’s was positive.</p><p>I further explored the meaning of the sentiment scores by looking at the exact sentences which scored the highest/lowest for each of the candidates.</p><p>Clinton’s most positive sentence has a score of 3.40.</p><blockquote>“Obviously, Medicare, which is a single-payer system, which takes care of our elderly and does a great job doing it, by the way, and then all of the people who were employed, but people who were working but didn’t have the money to afford insurance and didn’t have anybody, an employer or anybody else, to help them.”</blockquote><p>Trump’s most positive sentence has a score of 2.90.</p><blockquote>“But I want to do things that haven’t been done, including fixing and making our inner cities better for the African-American citizens that are so great, and for the Latinos, Hispanics, and I look forward to doing it.”</blockquote><p>Hillary’s most negative sentence has a score of -2.75.</p><blockquote>“There are children suffering in this catastrophic war, largely, I believe, because of Russian aggression.”</blockquote><p>Trump’s most negative sentence has a score of -2.50.</p><blockquote>“You know, when we have a world where you have ISIS chopping off heads, where you have — and, frankly, drowning people in steel cages, where you have wars and horrible, horrible sights all over, where you have so many bad things happening, this is like medieval times.”</blockquote><p>Interestingly, the most positive scoring sentences were generally optimistic outlooks of policies and views regarding the constituents, whereas the negative scoring sentences were regarding foreign hostilities.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-h45U6Aq9w4-Ylb8e1mOFA.png" /><figcaption>Scatterplot of Sentiment Scores (Blue is Clinton and Red is Trump)</figcaption></figure><h4>Conclusion</h4><p>Natural Language Processing and Sentiment Analysis are only <em>another</em> approach to evaluating the respective candidates based on the attitudes they have about America, the world, and life in general. For those like me who like to quantify and distill observations into whole truths, sentiment analysis offers metrics to compare the attitudes and evaluate rhetoric with a precise understanding. Based on the analysis of the second presidential debate, there are three conclusions that can be made:</p><ol><li>Trump used <strong>significantly</strong> more sentences than Clinton, which were often times <strong>incomplete</strong> and missing either a subject or predicate.</li><li>Trump was <strong>more</strong> <strong>negative</strong> than Clinton throughout the debate. However, Clinton said the most negative sentence in the debate.</li><li>The candidates used extremely positive rhetoric when discussing policies and plans for <strong>the American people</strong>, but they used extremely negative rhetoric when discussing <strong>war </strong>and<strong> foreign conflict</strong>.</li></ol><p>The method I used was rather naive, evaluating individual sentences rather than contextualizing. In order to fix problems like double negation in sentences and in larger in-context cases, another method I will explore is parsing sentences to build large binary trees. Based on the leafs and nodes of these trees, I can assign a score individually and use the handy power of multiplication between two nodes or leaves to eliminate the problem of double negations.</p><p>Credits to <a href="https://www.facebook.com/yash.pershad?fref=ts">Yash Pershad</a> for editing this mess and giving it a piercing title.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=16edb29d1c74" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>