<?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 Dhivya Rajprasad on Medium]]></title>
        <description><![CDATA[Stories by Dhivya Rajprasad on Medium]]></description>
        <link>https://medium.com/@dhivyakapil?source=rss-f0537fbf9e55------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/2*Idznmtz22qXcDNePFUqVXg.jpeg</url>
            <title>Stories by Dhivya Rajprasad on Medium</title>
            <link>https://medium.com/@dhivyakapil?source=rss-f0537fbf9e55------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 18:20:34 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@dhivyakapil/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[Building Recommender Systems for Retail E-Commerce — Part 1]]></title>
            <link>https://medium.com/@dhivyakapil/building-recommender-systems-for-retail-e-commerce-part-1-2aad2c99f2a3?source=rss-f0537fbf9e55------2</link>
            <guid isPermaLink="false">https://medium.com/p/2aad2c99f2a3</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[recommendation-system]]></category>
            <category><![CDATA[ecommerce]]></category>
            <category><![CDATA[personalization]]></category>
            <category><![CDATA[merchandising]]></category>
            <dc:creator><![CDATA[Dhivya Rajprasad]]></dc:creator>
            <pubDate>Tue, 22 Oct 2019 04:47:29 GMT</pubDate>
            <atom:updated>2019-10-22T04:47:45.740Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/300/1*uJSn34FItqJRoWlbR4tAPw.jpeg" /><figcaption>Recommender Systems</figcaption></figure><h3>Building Recommender Systems for Retail E-Commerce — Part 1</h3><p>Recommender systems have also become so essential in content discovery platforms that their importance cannot be emphasized more.</p><p>Their applications range from streaming platforms to music suggestions to shopping influencing our choices everyday.</p><p>Their importance in E-commerce is paramount as well as it reduces the friction of finding products suited to the users taste and increases the discoverability of products for the merchant making it a win-win situation for users and merchants alike.</p><p>Retail E-Commerce recommendation systems for established brands have the ability to increase the brand equity, influence shoppers and also personalize the experiences.</p><p>Building a recommender system is composed of many steps:</p><ol><li>Identifying the business need to be solved</li><li>Choosing the right data to build models</li><li>Researching the right algorithms</li><li>Online vs Offline learning decisions</li><li>Scalable architecture</li></ol><p><strong>BUSINESS NEED</strong></p><p>We will first discuss about identifying business need.</p><p>Business needs can have various forms and can help us define the problem with the algorithm choice.</p><p><strong>a) Pure Vanilla Experiences</strong></p><p>Problem: Users have difficulty in finding products that match their style</p><p>Solution: Build a recommendation carousel that can recommend products which are personalized to the user and the customer journey they are in</p><p><strong>b) Specialized Experiences</strong></p><p>Problem: Most people drop out at checkout page as they don’t reach free shipping and are frustrated to start the product discovery experience again.</p><p>Solution: Create a recommendation carousel which can recommend products that will help reach free shipping and might also complement the products in cart (not personalized to the user but to the products they have chosen in the current session)</p><p><strong>c) Brand Decisions</strong></p><p>Problem: Merchandising would like to suppress certain products (low priced items for high value purchasers) and promote certain products (moving repeat purchasers to try new products).</p><p>Solution: Recommendation Carousel is an ideal way to modularly influence consumers to purchase what makes the brand to revolutionize themselves</p><p><strong>d) Avoid Black Box Models</strong></p><p>Recommender systems are an ideal way to promote collaboration between several teams such as design to maintain the look and feel of the website, merchandising and site personalization to influence the products being shown (with minimal human interference to users preferences with concentration on brand image) and email teams to personalize content thus leading to increased cross collaboration instead of a black box model</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2aad2c99f2a3" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>