<?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 Muhammad Auwal Abubakar  on Medium]]></title>
        <description><![CDATA[Stories by Muhammad Auwal Abubakar  on Medium]]></description>
        <link>https://medium.com/@mauwal07?source=rss-464a2a6ebf20------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*5XdRdHRddvhMhjsMi1JW-A.jpeg</url>
            <title>Stories by Muhammad Auwal Abubakar  on Medium</title>
            <link>https://medium.com/@mauwal07?source=rss-464a2a6ebf20------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 13:26:06 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@mauwal07/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[DATA BIAS: AN ETHICAL CONSIDERATION IN DATA SCIENCE]]></title>
            <link>https://medium.com/@mauwal07/data-bias-an-ethical-consideration-in-data-science-54471ae91184?source=rss-464a2a6ebf20------2</link>
            <guid isPermaLink="false">https://medium.com/p/54471ae91184</guid>
            <dc:creator><![CDATA[Muhammad Auwal Abubakar ]]></dc:creator>
            <pubDate>Thu, 08 Feb 2024 22:49:07 GMT</pubDate>
            <atom:updated>2024-02-08T22:49:07.032Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>INTRODUCTION</strong></p><p>Data science is the process and method for extracting knowledge and insights from large volumes of disparate data. It is very clear from this definition that having correct data is key to making decisions that will have an impact on people’s lives. Consequently, this necessitates some ethical considerations about data to assist us in minimizing potential harms and unintended consequences from our data-driven actions. Some of these ethical considerations include dataset bias, data quality, data privacy, and intellectual property. However, this write-up focuses on dataset bias, specifically visual data (images and videos).</p><p><strong>DATASET BIAS</strong></p><p>Dataset bias (also called collection bias) is about selecting a non-representative subset of data for algorithm development, creating potential unfairness in result outcomes for diverse groups.</p><p>As stated earlier, this study focuses on visual data bias, and these are those that pertain to capture and collection. Visual data bias types include selection bias, framing bias, and label bias. These bias types are discussed below.</p><p>· Selection bias: Selection bias is the type of bias that occurs when individuals or groups in a study differ systematically from the population of interest, leading to a systematic error in an association or outcome. A more general definition is any disparity or association created as a result of the process by which subjects are included in a visual dataset.</p><p>· Framing bias: To frame is to select some aspects of a perceived reality and make them more salient in a communicating text in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation for the item described. A more general definition is any association or disparity that can be used to convey different messages and/or that can be traced back to how the visual content has been composed.</p><p>· Label bias is the bias that arises when labels differ systematically from the ground truth. A more general definition is any errors in the labeling of visual data with respect to some ground truth or the use of poorly defined or inappropriate semantic categories.</p><p>Figure 1 gives examples of selection, framing, and label bias, along with a list of applications that can be affected by each type of bias on the right.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/659/1*XX_E5vMMeE4DhSZGaBElVw.png" /><figcaption>Figure 1: Examples of selection, framing and label bias. On the right, a list of applications that can be affected by each type of bias.</figcaption></figure><p><strong>CONCLUSION</strong></p><p>In this write-up, the importance of accurate data in making data-driven decisions is presented. Also presented is dataset bias, which is one of the most important ethical considerations in data science. Specifically, visual dataset bias types, which include selection bias, framing bias, and labeling bias, are presented. However, as stated, there is no such thing as a bias-free dataset, so scientists and practitioners must become aware of the biases in their datasets and make them explicit.</p><p>For more interesting topics in data science, follow me @mauwal07</p><p><strong>Bibliography</strong></p><p>1. Coursera IBM Data Science professional course.</p><p>2. Fabbrizzi, S., Papadopoulos, S., Ntoutsi, E. &amp; Kompatsiaris, L. (2022). A survey on bias in visual datasets. <a href="https://doi.org/10.48550/arXiv.2107.07919">https://doi.org/10.48550/arXiv.2107.07919</a></p><p>3. <a href="https://github.com/Mauwal07/ArewaDS-Machine-Learning/blob/main/1-Introduction/02-ethics/README.md">https://github.com/Mauwal07/ArewaDS-Machine-Learning/blob/main/1-Introduction/02-ethics/README.md</a></p><p>Muhammad Auwal Abubakar</p><p>Potential Data Scientist</p><p>Fellow Arewa Data Science Academy</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=54471ae91184" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>