<?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 Sujit Gorantla on Medium]]></title>
        <description><![CDATA[Stories by Sujit Gorantla on Medium]]></description>
        <link>https://medium.com/@sujitgorantla?source=rss-fe4baca07b0f------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*MVO8vTFFDqjZ4n5L.</url>
            <title>Stories by Sujit Gorantla on Medium</title>
            <link>https://medium.com/@sujitgorantla?source=rss-fe4baca07b0f------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 16:03:12 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@sujitgorantla/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[Standard Size Plugin (E-commerce)]]></title>
            <link>https://medium.com/@sujitgorantla/standard-size-plugin-e-commerce-705c712e4ed0?source=rss-fe4baca07b0f------2</link>
            <guid isPermaLink="false">https://medium.com/p/705c712e4ed0</guid>
            <category><![CDATA[online-shopping]]></category>
            <category><![CDATA[ecommerce]]></category>
            <category><![CDATA[chrome-plugins]]></category>
            <dc:creator><![CDATA[Sujit Gorantla]]></dc:creator>
            <pubDate>Tue, 30 Oct 2018 13:16:43 GMT</pubDate>
            <atom:updated>2018-10-30T13:16:43.011Z</atom:updated>
            <content:encoded><![CDATA[<p>Category- Shoes</p><p><strong>Problem- </strong>Percentage of product returns due to size issue.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/576/1*cOWPTa9-hef_U03sd6xElg.jpeg" /></figure><p><strong>Product gap</strong>- The current user experience doesn&#39;t provide a standard size reference to the customers. Each brand maintains its own size standards leading to negative customer experience in terms of size /fit understanding. The current eCommerce platforms just provide a UK vs US sizing chart which has its own limitation of user understanding levels.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*480RTF9TnQAnqRLOvN2RUg.png" /></figure><p><strong>Feature</strong>- Provide a standard size reference chart for the complete platform . A chrome plugin which pop outs the standard size when you open a product page. It will also tell whether to go for a size up /down based on the your size provided in the profile while sign up</p><p><strong>Specifications-</strong></p><ol><li>The plugin should be allowed to capture the size Meta-tag from the flipkart Product description page (Like US 8 /UK 9 etc..)</li><li>The plugin should be allowed to capture the Brand name from the product description page</li><li>User should be able to store /save his preferred size ( Lets say UK 9 as his preferred shoe size)</li><li>The mapping of the sizes against the brands will be maintained in the back-end database</li><li>Based on the user size and the product size captured from the product page the plugin will suggest whether its a fit,or size up /down</li><li>The user can save these details for improving the further size reference experience</li></ol><p>The feature can be extended towards other categories such as T-shirts, Denims &amp; Innerwear</p><p><a href="https://footdistrict.com/en/size-chart-nike-new-balance-adidas">https://footdistrict.com/en/size-chart-nike-new-balance-adidas</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=705c712e4ed0" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Tackling Fake News using Emerging Technologies- 1]]></title>
            <link>https://medium.com/@sujitgorantla/tackling-fake-news-using-emerging-technologies-1-ef702aa08c7?source=rss-fe4baca07b0f------2</link>
            <guid isPermaLink="false">https://medium.com/p/ef702aa08c7</guid>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[fake-news]]></category>
            <dc:creator><![CDATA[Sujit Gorantla]]></dc:creator>
            <pubDate>Wed, 07 Mar 2018 10:49:20 GMT</pubDate>
            <atom:updated>2018-03-07T10:49:20.007Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*zYYb6jGT24OmJmsDOWwz5A.png" /><figcaption>Photo by FAKEBLOK</figcaption></figure><p>Few days back I was going through a Slate link which gives a detailed account of how Grncarov, a 20-year-old from Macedonian was able to trick Martina Navratilova, Serena Williams, the BBC and an entire Tennis community into believing his credentials in Tennis.</p><p>Link- <a href="https://slate.com/culture/2018/02/how-darko-grncarov-tricked-serena-williams-and-the-bbc.html">https://slate.com/culture/2018/02/how-darko-grncarov-tricked-serena-williams-and-the-bbc.html</a> )</p><p>At the same time i also came across a wired post on how Facebook might be moderating their news feed manually to push agendas of certain factions. There are several fake posts/news/articles that have been making rounds in social media channels sometimes even leading to severe public uproar. The way we have been moving forward its quite evident that in the coming years Fake News manufacturing will be at its pinnacle. So how do we the start taking control of this?</p><p>Fake news is a frequently used term these days made popular mainly after Modi’s Demonetization, US elections &amp; JIOs emergence. Facebook has become so popular that at one point in one of the South Asian countries it crossed google to become the most used medium to find latest news. Technological advancement has brought us this and we believe that the means to control its rise can also be provided by the technologies currently available.</p><p>I have divided this topic into two parts. The first part gives a basic understanding of how we can use current technologies to solve the fake content issues and we will function more like a base for my next part. In the second part we will look in depth at few Blockchain solutions that can be developed to tackle the fake news.</p><p><strong>Episode 1:</strong></p><p>Initially when we started, our idea was to tackle the issue of fake content at a website domain level. We researched and created an initial database of sites from where several fake articles /posts are spread. The idea was to create a chrome plugin which would pull the website domain address when the article URL is opened, check it against the database available and give a probability score of it being a fake article. The main idea was that the user could reach the article from any social media account or from any other source but he would finally end up at the parent website. But this process even though looks good in the initial phase had serious flaws at its core level.</p><ol><li><strong>Scalability issues.</strong> Updating the database frequently involves tiring manual work to research &amp; update</li><li><strong>Security issue</strong>- The idea was to keep an open sourced database of websites spreading the probable fake articles. Keeping it open sourced would keep the database to be updated continuously. There would be central team to moderate the database but it was still not a fool proof solution.</li><li><strong>Control issue</strong>- We might able to recognize a fake article if the website is in our database. In scenarios where some other website which is not a part of our database starts circulating those articles, it ends ups creating more confusion over an already confused fake article.</li><li>Using the domain address to check was also not an ideal way since it might create a wrong perception on articles which are actually not fake and it’s not ideal to penalize a whole website for missing out on one article.</li></ol><p>We felt it was an honest attempt from our end and paused the project to look at existing technologies that could help us solve these above issues and might help us in creating a business case out of this.</p><p><strong>Episode 2:</strong></p><p>The solution should take care that the open systems are still in place and at the same point use the intelligence and power of the readers to bring peace into this chaos</p><p>If we can pinpoint the author/source of the fake article or the website which first published it and keep them in front of all the readers, they automatically lose their credibility</p><p>It turns out that Blockchain technology does a pretty good job at this. I don’t want to bore you by explaining the Blockchain technology rather I will chalk out few features of it which might help us in tackling the problems we have seen before.</p><ul><li><strong>Consensus Based-</strong> A transaction on the Blockchain can be executed only when all the parties involved approve the transaction. Consensus can be reached by any of the popular mechanisms: Proof of work or Proof of stake (Credibility). Security issue which we were facing can be solved by using proof of stake consensus mechanism</li><li><strong>Immutability of the data.</strong> Once you have agreed on a transaction and recorded it, it can never be changed. You can subsequently record another transaction about that asset to change its state, but you can never hide the original transaction. This gives the idea of provenance of assets, which means that for any asset you can tell where it is, where it’s been and what has happened throughout its life.</li><li>It will be a public record for all of us so we know who said what and eventually whether things they claimed proved to be misleading.</li></ul><p>Now how do we do it?</p><p>Even though the ideal solution would be to pinpoint authors and the websites starting the fake articles. We believe it’s impossible to stop fake news manufacturing at the start but what we believe is, as a first step that if we can cut off their sources of propagation they will slowly become dormant and the next step would be to pinpoint the authors of these articles.</p><p>Basically since we want to keep our database be open sourced, it will rest on a block chain. New transactions (Fact checking) will be recorded on the Blockchain and the fact checkers will follow the proof of credibility mechanism to check the credibility of the articles by betting on their credibility scores.</p><p>At this point it’s important to know that in the initial phase the fact checker will be held accountable for the credibility of the article. The fact checkers will be rewarded with<strong> credibility scores </strong>when the article is correctly tagged as fake and will be penalized if it is wrongly tagged. When a story is published in the future, we can estimate its potential falsehood based on the fact checkers credibility score. Note that transactions mean that a request from the chrome plugin sent to the fake website identification engine (Explained in the later part of the article) to check the credibility of the website or article</p><p><strong>Block chain alone cannot solve these issues,</strong> it needs to integrate with data systems to continuously generate insights, for further scaling of the database and for improving the fake website identification engine.</p><p>The fake website identification engine does 3 things:</p><ol><li>The request from the plugin will be accepted by the engine. It will be checked against the existing database for credibility. If the record is not available or the confidence level is low (&lt; 70 percent), the transaction is published to the Blockchain for fact checking</li><li>Based on the transactions being generated the identification engine will start a rating for each website and the articles Accordingly the score is updated based on the credibility value generated after each transaction.</li><li>Also the engine will look for similar articles and websites that keeps popping up, a fake story might be presented in different forms in different articles when a certain article is flagged by the fact checker, the engine needs to check for similar stories and can classify them as fake or push it to the block chain again.</li></ol><p>To achieve all this <strong>machine learning </strong>will also be a key component along with the Blockchain technology.</p><p>Use of the <strong>ML components to directly classify &amp; flag the articl</strong>e based on the source and the author will take certain time since a large amount of data needs to consumed to train the systems to identity them precisely and maintain a credibility score of them. The initial scoring by the ML components might be risky and only when the probability scores of being fake are very high will the article be flagged.</p><p>Start with fact checkers to flag the articles by providing them an incentive in the form of credibility tokens and slowly build up the data systems to aid the fact checkers in making more informed decisions.</p><p><strong>In the next part of the article </strong>we will try to explore in depth about the few solutions which can be built to tackle fake news using Blockchain for example — <strong>Credibility protocols, Press trust Blockchain, Standalone Dapps</strong></p><p>P.S. Kindly provide your support to Fakeblok , <a href="https://fakeblok.com/">https://fakeblok.com/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ef702aa08c7" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>