<p><center><a target="_blank" href="https://www.datacamp.com/business/demo"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1603718763/banner_style_2_m573ef.png" /></a></p></p>
<p>Sudaman Thoppan Mohanchandralal, Regional Chief Data and Analytics Officer at Allianz Benelux, joined us to discuss the impact of building a data culture within an organization. If you want to learn more about Sudaman’s expertise in building data cultures, you can watch the <a href="https://www.datacamp.com/resources/webinars/building-data-cultures">full webinar</a>.</p>
<h2 id="why-is-it-important-to-build-a-data-culture-">Why is it important to build a data culture?</h2>
<p>When customer needs evolve, the business needs to evolve with it. Over the last decade, it has become easier for organizations to collect enormous amounts of data that can serve as the basis for innovative products, novel business strategies, and increased operational efficiencies. Nevertheless, the majority of organizations barely extract value from their data. The biggest challenges to creating a data-driven organization are not technical, they are cultural. Sudaman defines company culture as an aggregated and collective set of habits carried by the people within the organization. Culture design is not an easy task, it requires time and discipline to build organizational habits.</p>
<p>During the webinar, Sudaman explained how industries have shifted from mass-market to personalized services. For example, in the financial industry, the relationship with customers has evolved to a more personal experience, where customers are the center of an ecosystem of service providers that are creating personalized experiences for them. Data is the engine that powers this personalization, and building a data culture empowers organizations to maximize the value it extracts from it.</p>
<blockquote>
<p>“Data culture is not just an option to succeed in data analytics initiatives, it is business-critical” — Sudaman Thoppan Mohanchandralal</p>
</blockquote>
<h2 id="how-to-build-habits-that-lead-to-a-data-culture-">How to build habits that lead to a data culture?</h2>
<p>Building data culture starts with the reinforcement of habits and routines within the organization. Habits define the culture, and data-related habits and routines will transform the organization’s culture into a data culture.</p>
<p>Sudaman referred to the book <a target="_blank" href="https://www.goodreads.com/book/show/12609433-the-power-of-habit">“The Power of Habit”</a> by Charles Duhigg, which defines a habit as a three-element cycle. A habit starts with a cue or trigger, then a routine, and finally the reward. Every action in an organization is led by a specific reward and a trigger; however, the routine to obtain the reward can be changed to improve the organization’s habits.</p>
<center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1624556669/image_52_bor0fb.png" /></center>

<p>Sudaman mentioned that “the golden rule of habit change is to keep the cue, provide the same reward, but insert a new routine”. While evaluating the decision-making habits at Allianz Benelux, the Sales and Distribution Team realized that data was not as heavily used within the broker steering process. In this case, the routine was steering the broker, the reward was the business, and the cue was the timeframe and seasonality. The team was able to improve the cue and reward by implementing more data-driven decisions within the broker steering. More importantly, to change the routine and enable a high level of data culture, organizations must keep in mind three key elements:</p>
<ul>
<li><strong>Common language:</strong> the key to data culture success is to combine data literacy and subject matter expertise.</li>
<li><strong>Data governance:</strong> it is important to determine responsibilities for data ownership within the organization.</li>
<li><strong>Continuous measuring:</strong> building a data culture involves constant culture reassessment. Hence, organizations need to implement a continuous learning loop.</li>
</ul>
<h2 id="obstacles-towards-building-a-data-culture">Obstacles towards building a data culture</h2>
<p>There are two main hurdles that commonly stop companies from building a data culture.</p>
<ul>
<li><strong>Hurdle #1</strong> - Prioritization: besides understanding the importance of data culture, it is important to deal with the conflicting and unavoidable urgencies that pop up. Every organization has limited resources, hence, it is important for managers to know where data culture is in the list of priorities.</li>
<li><strong>Hurdle #2</strong> - Sustained Investments: data culture transitions take time and require constant reinforcement. Hence, it is important to maintain a consistent allocation of time and resources while the transition takes place.</li>
</ul>
<p>Before the organization decides to allocate time, effort, and money into building a data culture, it is important that they make sure to clarify three main factors.</p>
<ol>
<li><strong>Define your value:</strong> Why does an organization exist, and how does it maximize value for its customers?</li>
<li><strong>Measure deviation:</strong> What type of data is available and needed, and how can it be leveraged to further maximize value for customers?</li>
<li><strong>Appropriate re-prioritization and intervention:</strong> What should the organization do differently to develop the capabilities to extract value from data at scale?</li>
</ol>
<p>Finally, organizations must remember that achieving a data culture and data literacy within the organization is a continuous and iterative process.</p>
<h2 id="how-did-allianz-benelux-operationalize-its-culture-transformation-program-">How did Allianz Benelux operationalize its culture transformation program?</h2>
<p>While scaling out its culture transformation program, Allianz Benelux developed a data-driven decision-making (D3M) framework, designed to measure and evaluate the impact of the data culture. The D3M framework can be broken down into three main components:</p>
<ul>
<li><strong>Dimensions:</strong> these are universal in nature and measure the insights of data culture. Among these are data, strategy, technology, people, process, customer centricity, experimentation.</li>
<li><strong>Operationalization areas:</strong> These can be customized to the needs of the company based on the main areas of operation. The Allianz Benelux team determined the following: data infrastructure, leadership, tools, data literacy, embeddedness, customer centricity, experimentation.</li>
<li><strong>Subjects:</strong> subjects offer more granular insights around dimensions and operationalization areas. Among these are data availability, vision and strategy, usability, business integration, experimentation, customer-centricity.</li>
</ul>
<p>Sudaman explained how his team implemented the D3M framework by using surveys to derive scores for every dimension, operationalization area, and subject. By using this feedback loop, Allianz Benelux was able to determine the change in scores, and evaluate how and when to intervene in its culture change program. Through surveys, Allianz Benelux was also able to provide personalized learning combining external and internal resources from the global Data Analytics Academy for individuals to shore up their skill gaps, therefore elevating Allianz Benelux’s data culture and data fluency.  </p>
<h2 id="successful-ingredients-of-a-data-culture-program">Successful ingredients of a data culture program</h2>
<p>As mentioned by Sudaman, data culture starts with data literacy and skills. This is why the Data Analytics Academy partnered with DataCamp to roll out three custom learning programs with real-world data covering the gamut of data skills, from data literacy to data analysis and data science.  In just a few months,  more than 1000 Allianzers around the world are using DataCamp to improve and learn the fundamental skills needed to extract value from data at scale. This has allowed Allianz Benelux to maintain continuous reinforcement of their data culture and increase the rate of data literacy within the company.</p>
<p>In summary, to build a data culture, Sudaman recommends that organizations have a clear idea of the following:</p>
<ul>
<li><strong>Know your impact.</strong> How are we generating value for customers?</li>
<li><strong>Know the moves you need to make.</strong> What kind of decisions are different business functions making?</li>
<li><strong>Know your default behavior.</strong> What are the current habits of the organization?</li>
<li><strong>Reprogram redundant behavior.</strong> What habits are stopping the organization from adapting?</li>
<li><strong>Prioritize energy.</strong> develop a framework that fits the needs and priorities of the organization.</li>
<li><strong>Develop project lifecycle:</strong> monitor and measure ROI. Iterate every time it is needed.</li>
</ul>
<p>If you want to learn more about Sudaman’s insights on building data cultures, you can watch the <a target="_blank" href="https://www.datacamp.com/resources/webinars/building-data-cultures">full on-demand webinar</a>.</p>
<p><center><a target="_blank" href="https://www.datacamp.com/business/demo"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1603718763/banner_style_2_m573ef.png" /></a></p></p>


https://www.datacamp.com/community/blog/why-data-culture-matters

This blog post summarizes how and why data culture is a variable in any successful analytics endeavor, the meaning of data culture, and how it is an extension of organizational habits.

Why Data Culture Matters

<H2>Switch between modes in just one click</H2>

<p><center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1624384398/unified-navigration_wgv0rh.gif" /></center><br></p>
<p>Managing groups might be a time-consuming and tedious task, especially when it&#39;s an added responsibility at work. The new navigation aims to simplify the admin experience and minimize the time needed to complete your tasks.</p>
<p>Starting from June 9th, admins can access various groups by simply using a dropdown menu in the sidebar navigation. And we’re just getting started. In the coming weeks, we will enable a new feature allowing admins to switch between learning and group management modes in just one click.</p>
<H2>All your learning in one place </H2>

<p><center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1624384398/unified-navigation-2_nuote8.gif" /></center><br></p>
<p>Learning a new skill can be difficult, but it can be particularly challenging when you’re surrounded by constant distraction. To help you stay focused, we wanted to design an effective learning environment that encompasses all learning-related items you need to succeed. Over the next months, we will move all your assignments, custom tracks, and the leaderboard to the Learner Home.</p>
<H2>Drill down into individual performance </H2>

<p>Earlier this year we launched the Skill Matrix to give admins and managers an at-a-glance view of learners’ skill assessment scores and progress over time. To help you uncover even more insights, in the coming months we will make the Skill Matrix more robust by enabling a detailed view of learners’ progress and allowing admins to assign recommendations directly from the Skill Matrix.</p>
<p>We can’t wait to make these updates live. Our <a target="_blank" href="https://www.datacamp.com/community/blog/datacamp-2-0">mission</a> is to help people and organizations get ahead using the power of their data, and we are excited to make another step forward to achieving it with the new DataCamp navigation.</p>


https://www.datacamp.com/community/blog/meet-your-new-and-improved-datacamp-navigation

We’re excited to launch a new version of DataCamp navigation that makes it easier for you to achieve your learning goals—whether it’s closing one’s skill gaps or managing a group. 

Meet Your New and Improved DataCamp Navigation

<p>Click the image below to download our infographic on data literacy for responsible AI. For a deeper dive on how to spread the responsible use of AI with data literacy, explore our <a target="_blank" href="https://www.datacamp.com/resources/whitepapers/data-literacy-for-responsible-ai">exclusive guide</a>.</p>
<p><center><a target="_blank" href="https://s3.amazonaws.com/assets.datacamp.com/email/other/Data+Literacy+for+Responsible+AI+Infographic.pdf"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1624199048/Data_Literacy_for_Responsible_AI_Infographic_wibbvb.pdf" /></center></a></p>
<p><br><p><center><a target="_blank" href="https://www.datacamp.com/business/demo"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1603718763/banner_style_2_m573ef.png" /></a></p></p>


https://www.datacamp.com/community/blog/data-literacy-for-responsible-ai-infographic

Our infographic explains the importance of developing responsible AI and what it means for organizations today.

Data Literacy for Responsible AI [Infographic]

<p>DataCamp Digest is our newsletter aimed at providing the most up-to-date insights and news on all things data science. In this edition of the newsletter, we explore the future of jobs, discuss new tools and technologies, explore the future of AI regulation, and more.</p>
<p><center><a target="_blank" href="https://www.datacamp.com/groups/business"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1612888906/insights_qhc4kg.png" /></a></center></p> 

<h4 id="-the-future-of-jobs-in-the-era-of-ai-https-www-bcg-com-publications-2021-impact-of-new-technologies-on-jobs-boston-consulting-group"><a target="_blank" href="https://www.bcg.com/publications/2021/impact-of-new-technologies-on-jobs">The future of jobs in the era of AI</a> | Boston Consulting Group</h4>
<p>This report by BCG studies the impact of AI and automation on the job market in several key countries. It prescribes actions governments and organizations can take to alleviate automation woes, key amongst them being talent development.</p>
<h4 id="-new-ai-regulations-are-coming-are-you-ready-https-hbr-org-2021-04-new-ai-regulations-are-coming-is-your-organization-ready-harvard-business-review"><a target="_blank" href="https://hbr.org/2021/04/new-ai-regulations-are-coming-is-your-organization-ready">New AI regulations are coming, are you ready?</a> | Harvard Business Review</h4>
<p>The use of AI systems is becoming mainstream and regulators are setting guardrails to ensure the responsible use of AI. In this article, Andrew Burt outlines recent AI regulation efforts, and how organizations can prepare and ensure their models are compliant.</p>
<h4 id="-analytics-is-a-mess-https-benn-substack-com-p-analytics-is-a-mess-benn-stancil"><a target="_blank" href="https://benn.substack.com/p/analytics-is-a-mess">Analytics is a mess</a> | Benn Stancil</h4>
<p>Benn Stancil from Mode Analytics discusses the creative, messy, and non-linear nature of data science and analytics. If you ever feel stuck when solving a data science problem, remember that “A mess is the process of progress”.</p>
 <p><br><center><a target="_blank" href="https://www.datacamp.com/groups/business"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1612888907/Editor_s_Picks_nj17hh.png" /></a></center></p>

<h4 id="-openai-launches-100m-startup-fund-https-openai-com-fund-openai"><a target="_blank" href="https://openai.com/fund/">OpenAI launches $100M startup fund</a> | OpenAI</h4>
<p>OpenAI, the organization behind GPT-3, is launching a $100M startup fund for startups that leverage OpenAI tools specifically in “in fields where artificial intelligence can have a transformative effect—like health care, climate change, and education—and where AI tools can empower people by helping them be more productive.”</p>
<h4 id="-good-data-scientist-bad-data-scientist-https-ianwhitestone-work-good-ds-bad-ds-ian-whitestone"><a target="_blank" href="https://ianwhitestone.work/good-ds-bad-ds/">Good data scientist bad data scientist</a> | Ian Whitestone</h4>
<p>A short and snappy post that outlines what differentiates a good data scientist from a bad one.</p>
<h4 id="-speech-recognition-with-no-supervision-https-ai-facebook-com-blog-wav2vec-unsupervised-speech-recognition-without-supervision-facebook-ai"><a target="_blank" href="https://ai.facebook.com/blog/wav2vec-unsupervised-speech-recognition-without-supervision">Speech recognition with no supervision</a> | Facebook AI</h4>
<p>The team at Facebook AI leverages self-supervised learning to develop a speech recognition model that can learn to identify any language without any training data. This is huge for the recognition of languages where there is limited training data.</p>
<p><br><center><a target="_blank" href="https://www.datacamp.com/groups/business"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1612889346/Inspiration2_wxr25i.png" /></a></center></p>

<h4 id="-powerbi-coming-to-a-notebook-near-you-https-powerbi-microsoft-com-en-us-blog-announcing-power-bi-in-jupyter-notebooks-microsoft"><a target="_blank" href="https://powerbi.microsoft.com/en-us/blog/announcing-power-bi-in-jupyter-notebooks/">PowerBI coming to a notebook near you</a> | Microsoft</h4>
<p>Microsoft created a new Python package that lets you embed PowerBII directly in Jupyter Notebooks. This will make it much easier to embed compelling, editable data stories directly in Jupyter Notebooks.</p>
<h4 id="-lessons-on-ml-platforms-from-netflix-doordash-spotify-and-more-https-towardsdatascience-com-lessons-on-ml-platforms-from-netflix-doordash-spotify-and-more-f455400115c7-ernest-chan"><a target="_blank" href="https://towardsdatascience.com/lessons-on-ml-platforms-from-netflix-doordash-spotify-and-more-f455400115c7">Lessons on ML Platforms — from Netflix, DoorDash, Spotify, and more</a> | Ernest Chan</h4>
<p>Ernest Chan breaks down the components of the ML platforms that power the models of the most sophisticated data companies today.</p>
<h4 id="-microsoft-recommenders-https-github-com-microsoft-recommenders-microsoft"><a target="_blank" href="https://github.com/microsoft/recommenders">Microsoft Recommenders</a> | Microsoft</h4>
<p>A GitHub repository created by Microsoft that covers their best practices in creating recommender systems, with Jupyter Notebooks included. Check out the related projects section for similar repositories on NLP, computer vision, and more.</p>
<p><br><center><a target="_blank" href="https://www.datacamp.com/groups/business"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1612888937/Tools_and_Techniques_sf5hut.png" /></a></center></p>

<h4 id="-thinking-in-data-https-marketplace-visualstudio-com-items-itemname-dynamicwebpaige-thinking-in-data-paige-bailey"><a target="_blank" href="https://marketplace.visualstudio.com/items?itemName=dynamicwebpaige.thinking-in-data">Thinking in data</a>  | Paige Bailey</h4>
<p>If you use visual studio this one’s for you. Thinking in data is a set of VS Code extensions that supercharge your analytics and data science workflows on visual studio.</p>
<h4 id="-greykite-a-flexible-intuitive-and-fast-forecasting-library-https-engineering-linkedin-com-blog-2021-greykite-a-flexible-intuitive-and-fast-forecasting-library-reza-hosseini"><a target="_blank" href="https://engineering.linkedin.com/blog/2021/greykite--a-flexible--intuitive--and-fast-forecasting-library">Greykite: A flexible, intuitive, and fast forecasting library</a> | Reza Hosseini</h4>
<p>The data team at LinkedIn has open-sourced its Greykite forecasting library on Python. Check out the blog post to find out how it differs from other forecasting packages, and how the team applied it to forecasting problems at LinkedIn.   </p>
<h4 id="-flat-data-https-octo-github-com-projects-flat-data-github"><a target="_blank" href="https://octo.github.com/projects/flat-data">Flat data</a> | GitHub</h4>
<p>The team at GitHub released some tools that make it easy to work with and view data on GitHub.</p>
 <p><br><center><a target="_blank" href="https://www.datacamp.com/groups/business"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1612889508/From_DataCamp_c6kasv.png" /></a></center></p>

<h4 id="-webinar-building-data-cultures-https-www-datacamp-com-resources-webinars-building-data-cultures-"><a target="_blank" href="https://www.datacamp.com/resources/webinars/building-data-cultures">Webinar: Building Data Cultures</a></h4>
<p>In this webinar, regional Chief Data and Analytics Officer at Allianz Benelux Sudaman Thoppan Mohanchandralal deep dives into the ins and outs of building data cultures.</p>
<h4 id="-white-paper-your-organization-s-guide-to-data-maturity-https-www-datacamp-com-resources-whitepapers-your-organizations-guide-to-data-maturity-"><a target="_blank" href="https://www.datacamp.com/resources/whitepapers/your-organizations-guide-to-data-maturity">White Paper: Your Organization&#39;s Guide to Data Maturity</a></h4>
<p>In this guide, we outline a framework for evaluating, and scaling data maturity throughout the organization, define the various data maturity stages an organization goes through, and draw a path of initiatives organizations can take to achieve data fluency.</p>
<h4 id="-podcast-63-the-past-and-present-of-data-science-https-www-datacamp-com-community-podcast-the-past-and-present-of-data-science-"><a target="_blank" href="https://www.datacamp.com/community/podcast/the-past-and-present-of-data-science">Podcast: #63 The Past and Present of Data Science</a></h4>
<p>In this episode of DataFramed, Adel speaks with Sergey Fogelson, Vice President of Data Science and Modeling at Viacom on how data science has evolved over the past decade, and the remaining large-scale challenges facing data teams today.</p>
<p><br><center><a target="_blank" href="https://www.datacamp.com/business/demo"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1603718763/banner_style_2_m573ef.png" /></a></center></p>


https://www.datacamp.com/community/blog/digest-june-2021

Read our favorite articles from the last month.

DataCamp Digest June 2021: PowerBI in Jupyter Notebooks, the Future of Jobs, and More

<p>In college, I am planning to study computer science and economics. However, I did not want to wait to learn the basics of data science until I got to college. In that sense, DataCamp has truly helped me overcome the barriers of time and space.</p>
<br><center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1623766953/Screen_Shot_2021-06-15_at_10.21.57_AM_gqricw.png" /></center><br>

<p>I got access to <a target="_blank" href="https://www.datacamp.com/">DataCamp</a> as a member of Code for Nepal, a non-profit organization working to increase digital literacy and data literacy in Nepal. <a target="_blank" href="http://datacamp.com/donates">DataCamp Donates</a> is their <a target="_blank" href="https://www.datacamp.com/community/blog/datacamp-donates-50-partners">global effort</a> to build data skills and has donated thousands of subscriptions to organizations and schools globally.</p>
<p>As I worked to enhance the learning experiences of hundreds of others who were selected as Code for Nepal’s <a target="_blank" href="https://codefornepal.org/2021/02/14/launching-a-fellowship-program-to-build-data-skills-of-over-500-nepalis-globally">Data Fellows</a>, I was increasing my own knowledge of data science and familiarity with DataCamp. The courses on DataCamp are developed to meet the expectations of learners at every level—I have found them very helpful.</p>
<p>I had no prior experience with programming languages. I knew a little bit of HTML and Q-basic that I had learned in middle school. As a total beginner at the start of my DataCamp journey, I discovered Introduction to Python, Introduction to R, and Introduction to SQL and they were the perfect place to start.</p>
<p>The next step in my journey was having to decide whether I wanted to start with an individual course or a track. The courses started with the basics: types, variables, basic calculations. Therefore, even as a beginner that has never written or read a single line of code, I was able to follow the material presented very easily.</p>
<p>The courses on DataCamp take place on your browser so you don’t have to download anything to learn. I am very impressed with the concept of interactive e-learning with DataCamp as it allowed me to become active with my own learning experience. Interestingly, DataCamp XP is an experience point system that rewards learners like me for engaging in learning. XP is awarded to learners for progressing in courses, exercises, and projects.</p>
<p>When I use hints or show answer methods for solving problems a certain amount of my XP was used which motivated me to try and solve problems on my own as I did not want to lose the hard-earned XP.</p>
<p>DataCamp also has a section dedicated to getting real-life experience with projects. On the live events tab, there are always upcoming events like webinars and live training sessions to learn about data-driven organizations. I am thrilled to learn new things every day. DataCamp feels like a roadmap where I am in a data car that leads me safely by minimizing risk and providing clarity to reach my destination as a data scientist. This journey has to be my favorite of all time.</p>
<br><center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1623766953/Screen_Shot_2021-06-15_at_10.22.16_AM_iyf21n.png" /></center><br>

<p>All our DataCamp Donates fellows have had a positive experience learning different technologies like python, R, SQL, and Machine learning. In the era of the COVID-19 pandemic, as a high school student, I realized it is necessary to reinforce the various learning opportunities that prioritize strengthening my data skills and DataCamp was the best platform for it.</p>
<p>Finally, let me share that when I finished my first course, I was very proud of myself. After all, I managed to accomplish a significant step towards eventually becoming a data expert. I realized that becoming data fluent would require more time and dedication. I look forward to completing more courses, tracks, and projects that DataCamp has to offer.</p><br><br>
<p><i>Anshu recently graduated high school in Kathmandu, Nepal, and is matriculating to the University of Mary Washington in the fall. Code for Nepal’s mission is to help create a digitally and data-empowered Nepal together where everyone can thrive. Learn more about them by visiting <a target="_blank" href="http://codefornepal.org">their website</a>, where this post was <a target="_blank" href="https://codefornepal.org/2021/04/11/datacamp-experience.html">originally featured</a>. </i></p>
<br><center><a target="_blank" href="http://datacamp.com/donates"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1623777889/DCD_logo_regular_s7iazm.png" /></a></center>

<br><center><a target="_blank" href="http://codefornepal.org"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1623446367/code_for_nepal_3_zsmms5.png" /></a></center><br>


https://www.datacamp.com/community/blog/donates-stories-anshu-adhikari

A young DataCamp Donates scholarship recipient from Nepal shares her experience spending the last year learning.

How DataCamp Is Helping Me Learn Data Science for Free

<p>Welcome to the golden age of data. Today, companies have the privilege of collecting and storing ever-increasing amounts of data in all shapes and sizes. That’s great news, but how good are companies at extracting value from data? Not excellent, according to a <a target="_blank" href="https://www.accenture.com/_acnmedia/pdf-108/accenture-closing-data-value-gap-fixed.pdf">Forrester report</a> that cites that 60%to 73% of all data within an enterprise remains unused for analytics in 2019.</p>
<p>Why is that so? One of the reasons companies are stopping short of becoming fully data-driven, according to <a target="_blank" href="https://www.accenture.com/_acnmedia/PDF-115/Accenture-Human-Impact-Data-Literacy-Latest.pdf">Accenture</a>, is the lack of data skills that stifles workplace productivity. Surprisingly, in a survey of 9,000 companies, Accenture found that three-quarters of the employees feel overwhelmed or unhappy when working with data. Some employees report feeling as such because they are not ‘keeping up’ with the rapid changes in technological transformations. In fact, 59% of employees report feeling burnt out when using business intelligence and data analytics tools. Such unhappiness can be averted if employees are equipped with the skills required to deal with data, or, more precisely, data literacy.</p>
<p><a target="_blank" href="https://www.datacamp.com/tracks/data-literacy-fundamentals">Data literacy</a> can be defined as the ability to read, write, analyze, and communicate with data. Much like how one does not need to be Shakespeare to be considered literate, one does not need to become a data scientist to be data literate.  Data literacy is not about learning every single programming language or the most advanced data science skills. Instead, it is about understanding and making decisions with data.</p>
<p>Achieving data literacy is an essential step in a company’s roadmap to become data fluent. More concretely, it helps employees make <a target="_blank" href="https://www.datacamp.com/courses/data-driven-decision-making">data-driven decisions</a>, interact with data critically, form effective <a target="_blank" href="https://www.datacamp.com/resources/webinars/how-data-governance-enables-scalable-data-science">data governance</a>, and make ethical data decisions.</p>
<h2 id="data-literacy-is-the-keystone-to-making-data-driven-decisions">Data literacy is the keystone to making data-driven decisions</h2>
<p><a target="_blank" href="https://www.mckinsey.com/business-functions/mckinsey-analytics/our-insights/advanced-analytics-nine-insights-from-the-c-suite">McKinsey</a> found that employees are more likely to buy into a change when they understand it. Employees who understand both data science and its business application can spearhead the end-to-end development of use cases, and in turn influence others to make data-driven decisions.</p>
<p>For instance, the <a target="_blank" href="https://www.wsj.com/articles/the-houston-astros-a-case-study-of-the-transformative-power-of-analytics-1532102127">Houston Astros</a> used analytics to build its competitive advantage in a zero-sum industry, paving its way to victory in the Major League Baseline in 2017. Their secret? Executives who champion data literacy and make data-driven decisions. They employed data literate coaches (who can program in SQL) who could translate data insights to a language understood by players. Beyond that, the Astros built a strong data culture that permeates every facet of the team’s game strategy running the gamut from the process of player selection before the game to that of player position during the game. Such data-driven decisions would not have been possible without a data literate team.</p>
<h2 id="data-literacy-is-a-prerequisite-for-employees-to-interact-with-data-insights-meaningfully-and-critically-">Data literacy is a prerequisite for employees to interact with data insights meaningfully and critically.</h2>
<p>Data-literate employees are capable of creating and analyzing data visualizations that are integrated into the company&#39;s decision-making process. An example of such a company is PepsiCo, which uses <a target="_blank" href="https://www.datacamp.com/tracks/tableau-fundamentals">Tableau</a> and Hadoop to <a target="_blank" href="https://www.tableau.com/learn/webinars/how-pepsicos-big-data-strategy-disrupting-cpg-retail-analytics">visualize large volumes of data</a> that drive million-dollar sales decisions.</p>
<p>Moreover, being data literate is also a prerequisite to one’s ability to critically evaluate the validity of the data used to create visualizations.</p>
<blockquote>
<p>These employees who interrogate the reliability, correctness, and consistency of the data sources can potentially catch costly mistakes, thereby improving the company’s confidence and effectiveness in making data-driven decisions.</p>
</blockquote>
<h2 id="data-literacy-forms-the-foundation-of-effective-data-governance">Data literacy forms the foundation of effective data governance</h2>
<p><a target="_blank" href="https://www.datacamp.com/webinars/how-data-governance-enables-scalable-data-science">Data governance</a> is the set of policies, processes and organizational structure that define how the organization’s data is managed. It ensures that data is readily available, relevant to create value for the company, of high quality, and compliant with existing regulations. To build strong data governance, leadership must have basic data literacy to first understand the data context and needs of the organization before setting data policies.</p>
<p><a target="_blank" href="https://www.mckinsey.com/business-functions/mckinsey-digital/our-insights/designing-data-governance-that-delivers-value">McKinsey</a> suggested that the foundation for effective data governance involves a data organizational structure with three components:</p>
<ol><li>the data management office (DMO) in charge of defining policies and standards</li>
<li>domain leaders who set and execute domain-specific strategies</li>
<li>the data council that connects the domain leaders and DMO</li></ol>
<p>Conceivably, the leaders in each of these components should understand data processes sufficiently to set up unambiguous and fair data governance roadmaps. For instance, data-literate domain leaders who understand the value of the data can effectively work with the DMO to design and deploy an appropriate enterprise data lake for a domain.</p>
<h2 id="data-literacy-helps-companies-make-ethical-ai-decisions">Data literacy helps companies make ethical AI decisions</h2>
<p>Companies that deploy AI systems for real-world use cases face the risk of AI systems, such as <a target="_blank" href="https://arxiv.org/pdf/1606.06565.pdf">AI accidents</a>, breaches of data privacy, and <a target="_blank" href="https://hbr.org/2019/10/what-do-we-do-about-the-biases-in-ai">AI bias</a>. One way to mitigate such a risk is to engage technical and non-technical stakeholders to critically examine the processes and outputs of the data science system. Business stakeholders who are data literate play a crucial role in evaluating the risks of AI systems and ensuring that existing AI systems are fair and ethical.</p>
<p>For instance, as Facebook has come under intense scrutiny in recent years over its recommendation systems, the <a target="_blank" href="https://ai.facebook.com/blog/what-ai-fairness-in-practice-looks-like-at-facebook/">Facebook Responsible AI Initiative</a> was founded. It consists of a multidisciplinary team of not just data scientists, but also philosophers and social scientists. This team evaluates existing systems and determines ethical standards openly. These qualitative standards are then translated and implemented by data scientists as quantitative metrics. Such debates on AI risks are productive only when all the stakeholders are well-informed and have a minimum level of data literacy to weigh the cost and benefits of an AI system. Similarly, organizations looking to operationalize and govern AI need a multidisciplinary team of business and data experts that have a common data language.</p>
<h2 id="building-organizational-data-literacy">Building Organizational Data Literacy</h2>
<p>Clearly, inculcating a strong data literacy in all employees brings about a myriad of benefits. Data literacy programs take many forms, including existing skills initiatives, e-learning courses, or specialized classroom training. Unsurprisingly, designing in-house data literacy programmes is a non-trivial and long-drawn process that requires meticulous planning. This is whyDataCamp for Business provides an interactive learning platform for companies that need to upskill and reskill their people on data skills. With topics ranging from data literacy, and data science to data engineering and machine learning, over 1,600 companies trust DataCamp for Business to upskill their talent.</p>
<br><p><center><a target="_blank" href="https://www.datacamp.com/business/demo"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1603718763/banner_style_2_m573ef.png" /></a></p>

https://www.datacamp.com/community/blog/why-data-literacy-is-important-for-your-team

In this blog post, learn why data literacy is important and why some companies aren't extracting value from their data like they should be.

Why Data Literacy Is Important for Your Team

https://www.datacamp.com/community/blog/creating-trust-in-data-with-data-observability

In this episode, Adel speaks with Barr Moses, CEO, and co-founder of Monte Carlo, on the importance of data quality and how data observability creates trust in data throughout the organization.

Creating Trust in Data with Data Observability

<p>Over the last few years, AI has seen significant growth in utilization and effectiveness in organizations. Its applications range from predictive analytics for more efficient and accurate business operations, to machine vision for direct interactions with the world, and natural language processing (NLP) for more effective interaction with technology through language. <a href="https://www.accenture.com/us-en/insights/artificial-intelligence/ai-investments">Accenture reports</a> that 84% of surveyed executives believe they need to leverage AI to achieve their growth objectives.</p>
<p>With this fast growth in a relatively new and maturing technology comes a lot of responsibility and ethical concerns. <a target="_blank" href="https://www.capgemini.com/news/ai-and-ethics-2020-report/">Capgemini found</a> that 70% of customers expect transparency and fairness in their AI interactions. Additionally, regulators are beginning to tighten regulations around AI transparency and fairness. The European Union recently released a set of AI regulations that aim to minimize the harmful impacts of AI systems, and the FTC announced plans to go after companies selling biased algorithms (<a target="_blank" href="https://www.technologyreview.com/2021/04/21/1023254/ftc-eu-ai-regulation-bias-algorithms-civil-rights/">MIT Tech Review</a>).</p>
<p>The increases in ethical concerns are justified and driven by AI use cases that increasingly involve highly sensitive settings such as healthcare, judicial systems, and finance. Without proper tuning and understanding of the potential ethical risks, AI can lead to the reproduction and amplification of historical bias and discrimination. For example, <a target="_blank" href="https://www.propublica.org/article/how-we-analyzed-the-compas-recidivism-algorithm">ProPublica investigated the COMPAS algorithm</a> which was used to inform sentencing for criminals by predicting the likelihood of an individual to re-offend. They discovered that black defendants were significantly more likely to receive false-positive recommendations suggesting they would re-offend when they would not while white defendants were more likely to receive false-negative recommendations suggesting they would not re-offend. </p>
<p>Similarly, OpenAI’s <a target="_blank" href="https://github.com/openai/gpt-3">GPT-3</a> NLP transformer model with 175 billion parameters has demonstrated significant bias in generating sentences that include <a target="_blank" href="https://arxiv.org/pdf/2101.05783v1.pdf">religion</a>. The model has also shown bias toward other groups, including gender and race which is discussed in the <a target="_blank" href="https://arxiv.org/pdf/2005.14165.pdf">original paper</a>.  </p>
<p>As AI becomes more ubiquitous, it’s crucial to ensure the responsible use of AI. If we are to trust these systems to inform important and sensitive decisions, they cannot be biased or include historical stereotypes.</p>
<h2 id="what-is-algorithmic-bias-">What is Algorithmic Bias?</h2>
<p>Algorithmic bias is pretty prominent in the news, as seen in <a target="_blank" href="https://www.reuters.com/article/us-amazon-com-jobs-automation-insight/amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK08G">gender bias</a> in hiring decisions favoring men, online advertisements that reproduce redlining on a map (<a target="_blank" href="https://www.theguardian.com/technology/2019/mar/28/facebook-ads-housing-discrimination-charges-us-government-hud">housing discrimination</a>), and healthcare algorithms that include <a target="_blank" href="https://science.sciencemag.org/content/366/6464/447">racial bias</a>. All of these examples clearly show a group being treated differently by the model than another group.</p>
<p>Many metrics exist for formally defining algorithmic bias. The basic definition is that an algorithm treats groups in the data differently. The groups we are concerned about, such as race, gender, religion, and pregnancy or veteran status are called protected or sensitive characteristics. When analyzing an algorithm, it’s necessary to understand what groups are potentially most vulnerable to harmful impacts of the model.</p>
<p>There are two categories of metrics to determine the bias and fairness of an algorithm: fairness by representation and fairness by error.</p>
<h3 id="fairness-by-representation">Fairness by Representation</h3>
<p>These metrics evaluate the outcomes of the model for different groups. For example, these metrics could look at the proportion of men who get called back for an interview versus women for a hiring algorithm.</p>
<h3 id="fairness-by-error">Fairness by Error</h3>
<p>Fairness by error metrics observe the difference in the proportion of error rates made for each group. For example, in the COMPAS algorithm bias described above, the false-positive rate for black defendants was significantly higher than other groups, and the false-negative rate for white defendants was significantly lower.</p>
<h3 id="causes-of-algorithmic-bias">Causes of Algorithmic Bias</h3>
<p>Bias in an algorithm is caused by the underlying data. Simple models like linear regression and more complex transformer models are not biased on their own; they produce outputs based on the data provided to them. Below are five causes of bias in data.</p>
<ol>
<li><strong>Skewed dataset:</strong> A dataset may have different proportions of data for each group or training examples for each group which may lead to biased model outputs.</li>
<li><strong>Tainted examples:</strong> The data may have unreliable labels or historical bias may exist in the data.</li>
<li><strong>Limited Features:</strong> The collected features may be unreliable or not informative for specific groups due to bad data collection practices.</li>
<li><strong>Samples Size:</strong> AI algorithms struggle to perform well with smaller datasets.</li>
<li><strong>Proxy Features:</strong> Some features can leak information about protected variables, even if the protected variable is not included in the data. <a target="_blank" href="https://arxiv.org/pdf/1707.08120.pdf">Research has shown</a> that variables like zip code, sports activity, and the university a person attended can all indirectly inform a model about protected variables which can lead to algorithmic bias.</li>
</ol>
<h2 id="what-are-mitigation-techniques-that-can-be-used-">What are mitigation techniques that can be used?</h2>
<p>We now have a good formal understanding of what algorithmic bias is and what it is caused by. There are two potential avenues for mitigating algorithmic bias: technical techniques and AI governance. Most likely a combination of these two techniques will yield the best results in reducing the presence of bias in an algorithm.</p>
<h3 id="technical-mitigation-techniques">Technical Mitigation Techniques</h3>
<p>Within technical techniques, there are three approaches: pre-processing techniques which reduce bias in the original data before training, in-processing methods that mitigate bias during model training, and post-processing algorithms that reduce bias by modifying predictions.</p>
<p><a target="_blank" href="https://core.ac.uk/download/pdf/81728147.pdf">This paper</a> describes four common pre-processing techniques used in the literature in the introduction:</p>
<ul>
<li><strong>Suppression:</strong> remove a variable that correlates most with a sensitive variable</li>
<li><strong>Massaging:</strong> Relabel some observations in the data in an efficient way using a ranker to remove discrimination in the input data</li>
<li><strong>Reweighting:</strong> Assign weights to each tuple in the training data to remove discrimination</li>
<li><strong>Sampling:</strong> Sample from each group to get a more balanced representation of each group using either uniform sampling or preferential sampling by duplicating or skipping observations.</li>
</ul>
<p>Other techniques like <a target="_blank" href="https://krvarshney.github.io/pubs/CalmonWVRV_nips2017.pdf">this paper’s</a> probabilistic formulation and <a target="_blank" href="http://sorelle.friedler.net/papers/kdd_disparate_impact.pdf">this paper</a>, which describes two “repair procedures,” provide good methods of removing bias as well in the pre-processing stage.</p>
<p>In-processing techniques focus on removing bias during training. These techniques typically include fairness constraints which impact the optimization problem and lead to more fair results. Examples discussing in-process methods include this one looking at <a target="_blank" href="https://arxiv.org/pdf/1806.06055.pdf">fairness constraints</a> and this one looking at <a target="_blank" href="https://www.kamishima.net/archive/2012-p-ecmlpkdd-print.pdf">fairness-aware classification</a>.</p>
<p>Finally, post-processing algorithms modify model outputs to reduce bias. These algorithms can either change predictions for privileged or unprivileged groups as seen by <a target="_blank" href="https://arxiv.org/pdf/1709.02012.pdf">Pleiss et al.</a> and <a target="_blank" href="https://arxiv.org/pdf/1610.02413.pdf">Hardt et al.</a> or they reduce bias when an algorithm output is close to the decision boundary (where it is less sure about the result) as seen in <a target="_blank" href="https://mine.kaust.edu.sa/Documents/papers/ICDM_2012.pdf">this paper</a> on decision theory.</p>
<p>For any of these techniques, it is essential to discuss with all stakeholders involved the trade-offs in accuracy and the risks being taken by changing the underlying data or model results. These techniques can lead to predicting outcomes that have not been seen historically or are not seen in the data.  </p>
<h3 id="governance-and-accountability">Governance and Accountability</h3>
<p>There are a lot of benefits to using AI in organizations. However, as seen throughout the blog post, ethical risks posed by AI systems lead to organizational risks such as reputational damage, which lead to loss of revenue. No firm wants to have a headline saying they used a biased algorithm to make business decisions that negatively impacted a group of customers or employees.</p>
<p>Proper AI governance is conducted through comprehensive risk management. The specific use case of the AI decision and the associated impact assessment for each stakeholder needs to be identified. Understanding the potential monetary and human risks associated with the use case is important. Once the magnitude of each risk is accurately assessed for each stakeholder, the likelihood for these repercussions should be calculated.</p>
<p>AI governance best practices include comprehensive checklists for stakeholders, automated and manual testing, business standards compliance testing, detailed reports, and assertions from individuals and groups who are directly responsible for the model. For more, details make sure to download our white paper.</p>
<h2 id="responsible-ai-relies-on-data-literacy">Responsible AI Relies on Data Literacy</h2>
<p>Scaling responsible AI in organizations requires the development of data literacy within the organization. Successful AI governance includes multiple stakeholders engaging with and adopting a single framework. Understanding the scope of projects, how they are deployed and governed, and their impact and potential risks is a requirement for responsible AI for all involved stakeholders. <a target="_blank" href="https://www.accenture.com/_acnmedia/PDF-92/Accenture-AFS-Responsible-AI.pdf">Accenture argues</a> that data literacy builds trust and allows for responsible AI by including diverse perspectives which lead to consistently better outcomes.</p>
<p>By increasing baseline data literacy in an organization, subject matter experts who do not consistently interact with AI systems can include their expertise in these systems and communicate with technical stakeholders more effectively. This improved communication leads to more trust throughout the organization for all stakeholders involved.</p>
<p>The important aspects of data literacy for promoting responsible AI in an organization include an understanding of the following areas: the data science and machine learning workflow, different data roles, the flow of data through an organization, the distinction between various types of AI systems, and evaluation metrics for machine learning models.</p>
<h2 id="increase-baseline-data-literacy-to-scale-responsible-ai">Increase Baseline Data Literacy to Scale Responsible AI</h2>
<p>Training all stakeholders in an organization is required to promote responsible AI. Avoiding the associated risks with algorithmic bias is necessary to allow an organization to successfully take advantage of the many benefits of leveraging AI in business contexts. Data literacy is the key to successful communication between stakeholders, trust within the organization, and diverse solutions to be explored for responsible AI. The EU is allocating €1.1 trillion to their <a target="_blank" href="https://ec.europa.eu/social/BlobServlet?docId=22832&amp;langId=en">Skills Agenda</a>, with a heavy emphasis on AI skills and data literacy for regulators, demonstrating the importance of these tools in organizations. Many programs teach the fundamentals of data literacy. For more details on how data literacy can fuel responsible AI — check out our latest white paper.  </p>
<br><p><center><a target="_blank" href="https://www.datacamp.com/business/demo"><img src="https://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1603718763/banner_style_2_m573ef.png" /></a></p>


https://www.datacamp.com/community/blog/the-case-for-responsible-ai

We recently released a report co-written by DataRobot’s VP of Trusted AI Ted Kwartler, Global AI Ethicist Haniyeh Mahmoudian, and Managing Director of AI Ethics Sara Khatry. Here’s a run-down of what to expect.

community

datacamp

Official Blog