<?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 Odunayo Aliu on Medium]]></title>
        <description><![CDATA[Stories by Odunayo Aliu on Medium]]></description>
        <link>https://medium.com/@odunthedatagirl?source=rss-42bca778fa7e------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*PAGeTq3LnMryjdFEt3fl4g@2x.jpeg</url>
            <title>Stories by Odunayo Aliu on Medium</title>
            <link>https://medium.com/@odunthedatagirl?source=rss-42bca778fa7e------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 10:08:02 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@odunthedatagirl/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[Enterprise Data Governance Analysis]]></title>
            <link>https://medium.com/@odunthedatagirl/enterprise-data-governance-analysis-770b8a3cf95f?source=rss-42bca778fa7e------2</link>
            <guid isPermaLink="false">https://medium.com/p/770b8a3cf95f</guid>
            <dc:creator><![CDATA[Odunayo Aliu]]></dc:creator>
            <pubDate>Thu, 05 Feb 2026 08:20:03 GMT</pubDate>
            <atom:updated>2026-02-05T08:20:03.235Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IiQZrgeentHMmj1KzUa7-Q@2x.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1SsSoj94MWCSeFQhyiGhuw@2x.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QKEfUZPBE9Wlalf8XRy7Rg@2x.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9RHdqXnKuWaKMUGcFdr1GA@2x.jpeg" /></figure><h4>Page 1: Executive Overview</h4><p>Page Overview</p><p>This page provides a high-level summary of the organization’s data governance health, focusing on the relationship between data quality and business impact across various domains. It tracks overall compliance metrics, data quality scores, and estimated financial risks.</p><p>Findings</p><p>Financial Risk: The organization faces a total estimated fine of €553,057,673 based on 10,171 governance findings.</p><p>Data Quality: The overall Data Quality Score is 93.94, with Accuracy at 93.08% and Completeness at 94.19%.</p><p>Domain Maturity: Marketing has the lowest Governance Maturity Score (2.17) despite having the highest number of audit findings (2,152) and the largest estimated fine (€123,103,697).</p><p>Impact vs. Quality: The scatter plot indicates that Marketing has a high Business Impact Score but significantly lower Data Quality compared to Finance and IT.</p><p>Recommendations</p><p>Prioritize Marketing Remediation: Launch a targeted data cleanup for the Marketing domain, as it represents the highest financial risk and the lowest maturity level.</p><p>Focus on Accuracy: Implement automated validation rules to bridge the gap between Accuracy (93.08%) and Completeness (94.19%) to reduce total findings.</p><p>Audit Frequency: Review domains like Finance, which has a high “Days Since Last Audit” count (249,576), to ensure records remain compliant over time.</p><p><strong>Page 2: Governance Findings</strong></p><p>Page Overview</p><p>This page focuses on the operational details of data audits, including geographic distribution of issues, audit trends over time, and individual accountability through Data Stewards.</p><p>Findings</p><p>Regional Hotspots: The central European region shows the highest concentration of issues, with one area alone accounting for 5,580 findings.</p><p>Steward Performance: Compliance Risk Scores vary significantly by steward, with Marketing Analytics and Platform Engineering overseeing high-risk areas (scores of 61 and 58 respectively).</p><p>Critical Records: Specific records, such as Access_Logs, are identified as high-risk with 10 findings and a Compliance Risk Score of 97.60.</p><p>Classification Balance: Data is evenly split between Public, Confidential, and Internal classifications (33.33% each).</p><p>Recommendations</p><p>Geographic Resource Allocation: Deploy additional governance resources to the central European region to address the 5,580 open issues.</p><p>Targeted Steward Support: Provide the Marketing Analytics steward with more support or training, as their assigned data currently carries the highest compliance risk.</p><p>High-Risk Remediation: Immediately address “Record 3&quot; (Access_Logs) due to its extremely high Compliance Risk Score of 97.60 to prevent potential breaches.</p><p>Page 3: Retention &amp; Cost Tracker</p><p>Page Overview</p><p>This page tracks the financial exposure related to specific regulations and evaluates the efficiency of the organization’s remediation efforts.</p><p>Findings</p><p>Regulation Exposure: GDPR represents the largest financial threat, accounting for €1,226,928,588 of the total potential fines.</p><p>Efficiency Gap: The Issue Closure Efficiency is currently low at 34.25%, contributing to a high remediation cost of over €276 million.</p><p>Retention Variance: Retention policies vary by domain, with Operations holding data for 5.09 years on average, while HR holds it for only 3.54 years.</p><p>Fine Trends: The estimated fine trend is projected to remain high through 2024 and 2025, with quarterly exposures often exceeding €230M.</p><p>Recommendations</p><p>Improve Closure Rates: Focus on increasing the Issue Closure Efficiency (34.25%) to reduce the overall remediation cost and fine exposure.</p><p>Prioritize GDPR Controls: Since GDPR-related risks make up the bulk of potential fines, prioritize these audits over Internal Policy or NIS2 findings.</p><p>Standardize Retention: Audit the Operations domain to determine if the 5.09-year retention period is legally required or if reducing it could lower financial risk.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=770b8a3cf95f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Healthcare Data Analysis Using Power BI]]></title>
            <link>https://medium.com/@odunthedatagirl/healthcare-data-analysis-using-power-bi-967602277dca?source=rss-42bca778fa7e------2</link>
            <guid isPermaLink="false">https://medium.com/p/967602277dca</guid>
            <category><![CDATA[power-bi]]></category>
            <category><![CDATA[healthcare-technology]]></category>
            <category><![CDATA[dataviz]]></category>
            <category><![CDATA[data-analysis]]></category>
            <category><![CDATA[data-visualization]]></category>
            <dc:creator><![CDATA[Odunayo Aliu]]></dc:creator>
            <pubDate>Mon, 19 Jan 2026 11:00:25 GMT</pubDate>
            <atom:updated>2026-01-29T17:20:44.731Z</atom:updated>
            <content:encoded><![CDATA[<p>In this project, I explored healthcare data spanning from January 2019 to December 2024 to uncover critical patterns in patient demographics, hospital operations, and financial performance.</p><p>By analyzing a dataset of 55,500 total admissions, I discovered a remarkably balanced distribution of admission types, with Elective (33.61%), Urgent (33.47%), and Emergency (32.92%) cases each representing roughly one-third of the hospital’s volume. A key operational finding was the average length of stay of 15.51 days, which remained consistent across the six years, suggesting a stable but high demand for inpatient resources.</p><p>From a clinical perspective, I identified that 33.56% of test results were “Abnormal,” slightly outnumbering “Normal” and “Inconclusive” findings. The data also revealed that the average patient age is approximately 51.5 years, with a significant volume of cases involving chronic conditions like Diabetes, Obesity, and Arthritis. The report presents a strategic financial and clinical audit of healthcare operations from 2019 to 2024. With an annual revenue exceeding $281 million in the 2023 calendar year alone, the analysis highlights the high-value impact of chronic disease management specifically Obesity, Diabetes, and Hypertension on the bottom line. By analyzing the $25,574 average bill per patient against clinical outcomes and diagnostic trends, this analysis identifies opportunities for revenue cycle optimization and cost-effective patient care models.</p><h3>Insights from the Analysis:</h3><h4>Patient Demographics &amp; Admissions</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PWa9sdkF9HvmJi2_VyOROQ.jpeg" /><figcaption>Page 1</figcaption></figure><ul><li>High Volume and Balanced Admission Types: There were 55,500 total admissions between January 2019 and December 2024. These are distributed almost equally across admission types: Elective (33.61%), Urgent (33.47%), and Emergency (32.92%).</li><li>Aging Patient Population: The average patient age is 51.37 years. The data shows that admissions are fairly consistent across middle-to-older age groups, peaking slightly in the late 40s to early 60s.</li><li>Abnormal Test Results: A significant 33.56% of all test results are “Abnormal,” with 18,627 patients receiving abnormal findings compared to 18,517 “Normal” and 18,356 “Inconclusive”.</li></ul><h4>Operational Efficiency</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uLMN0OO_lg6vePq2gt0DEA.jpeg" /></figure><ul><li>Length of Stay: The average length of stay is 15.51 days. This is relatively high for a general hospital setting, suggesting a high volume of complex cases or potential bottlenecks in the discharge process.</li><li>Resource Utilization: The hospital operates with 400 available rooms and maintains a staff of 40,341 doctors, indicating a very high doctor-to-patient ratio or a large network of affiliated providers.</li></ul><h4>Financial Performance</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TPgIgNtn6UpU1M5lsWvwbQ.jpeg" /></figure><ul><li>Revenue Concentration: Total revenue for the 2023 calendar year reached $281,977,253.</li><li>High-Cost Medical Conditions: Revenue is heavily driven by chronic conditions. For example, Obesity ($238.2M), Diabetes ($238.5M), and Hypertension ($235.7M) are the top revenue generators.</li><li>Billing Per Patient: The average bill per patient is $25,574, which aligns with the high average length of stay.</li></ul><h3>Recommendations</h3><h4>Operational Recommendations</h4><ul><li>Optimize Patient Flow: Implement a virtual queue or triage system to manage the high volume of emergency and urgent admissions (over 66% combined) to reduce physical crowding and improve patient experience.</li><li>Reduce Length of Stay (LOS): Review discharge protocols to lower the current average of 15.5 days. Strategies like pre-visit planning and standardized care transitions can help move patients safely into community care or home settings faster, freeing up beds.</li><li>Address Room Utilization: With 400 rooms and over 11,000 patients annually (in the 2023 subset), the hospital should use predictive analytics to forecast peak admission times and prevent “boarding” in the emergency department.</li></ul><h4>Financial &amp; Revenue Cycle Recommendations</h4><ul><li>Target Chronic Condition Management: Since Diabetes, Obesity, and Hypertension drive the bulk of revenue, the hospital should invest in specialized outpatient clinics or “Health Wellness Programs” to manage these patients preventatively, which can improve long-term outcomes while stabilizing revenue streams.</li><li>Improve Billing Accuracy: To maintain the $25,574 average bill, ensure that Charge Capture Specialists are auditing records for complex stays to prevent “revenue leakage” from unbilled supplies or services.</li><li>Insurance Verification: Given the high billing amounts, implement real-time insurance eligibility verification at registration to reduce claim denials and ensure patients understand their financial responsibility early.</li></ul><h4>Clinical Quality Recommendations</h4><ul><li>Inconclusive Result Follow-up: With nearly 18,356 inconclusive results, the hospitals should establish a protocol for automatic re-testing or specialist review to ensure these do not lead to missed diagnoses or unnecessary readmissions.</li><li>Preventative Screenings: The high rate of abnormal results (33.56%) suggests a need for more robust community outreach for early screening, particularly for prevalent conditions like Cancer and Hypertension identified in the data.</li></ul><p>You can explore the interactive dashboard <a href="https://app.powerbi.com/groups/me/reports/f64b8397-d4ab-40e5-b8b2-86f6a8f27b82/7b34c23f971722b6630e?experience=power-bi">here</a></p><p>Thank you!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=967602277dca" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Employee Performance Analysis]]></title>
            <link>https://medium.com/@odunthedatagirl/employee-performance-analysis-733b4bed2cc5?source=rss-42bca778fa7e------2</link>
            <guid isPermaLink="false">https://medium.com/p/733b4bed2cc5</guid>
            <category><![CDATA[power-bi-dashboard]]></category>
            <category><![CDATA[zoomcharts]]></category>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[hr-data-analytics]]></category>
            <category><![CDATA[data-analysis]]></category>
            <dc:creator><![CDATA[Odunayo Aliu]]></dc:creator>
            <pubDate>Thu, 18 Sep 2025 11:39:00 GMT</pubDate>
            <atom:updated>2025-09-19T10:23:58.094Z</atom:updated>
            <content:encoded><![CDATA[<p>I assumed the role of an HR data analyst for an FP20 challenge 30 and my task was to draw out insights using employee performance records.</p><p>The report is organized into three main sections: Overview, Performance, and Outcome, offering a view of the employee lifecycle.</p><p>Insights reveal trends in salary, performance, and turnover, with a strong emphasis on identifying top performers and high-turnover areas.</p><p><strong>Employee Overview</strong></p><p>This page provides a high-level summary of the workforce.</p><p>- Headcount &amp; Attrition: The company has a total of 7,500 employees, with an active workforce of 6,009 (80.1%) and 1,491 who have exited (19.9%). The Store Operations department has the highest number of exited employees, indicating a potential area of concern for retention.</p><p>- Compensation: The average salary is $27.25K, with an average bonus of $248.60. Salary varies significantly by job level, with Executives earning an average of $111.15K, while Entry level employees earn $23.01K.</p><p>- Salary Increase Flag shows that only a small percentage of employees received a bonus or salary increase, with the majority falling into the “false” category (98.64%). This suggests that salary increase is not a widespread practice within the company.</p><p><strong>Performance Trends</strong></p><p>This page explains the performance-related metrics over time and by department.</p><p>- Performance Rating: The average performance rating fluctuates throughout the year, but generally remains above 3.5 on a scale of 5, indicating a solid overall performance from the workforce. The highest ratings were observed in July 2024.</p><p>- Training &amp; Performance Correlation: There is a positive correlation between Training hours and performance according to the scatter plot. Employees who have received more training hours tend to have higher performance ratings, suggesting that training directly contributes to employee success.</p><p>- Manager Performance: Anna Love and David Gutierrez are the top performing managers.</p><p><strong>Business Outcome</strong></p><p>This page connects employee performance to business results, particularly sales.</p><p>- Sales Performance: The dashboard tracks actual sales against targets for individual stores. The Stores above Target metric shows that 150 stores have exceeded their goals.</p><p>- Sales vs. Target Analysis: A detailed table shows that the New York Superstore #81 has the highest sales, exceeding its target by 2.95%. The “% Diff” column is a key metric for quickly identifying which stores are over or under their goals.</p><p>- Satisfaction by Department: Average Employee Satisfaction Score by Department chart shows that Store Operations has the highest satisfaction score at 7.32, even though it has a high number of exits, which may indicate a complex set of factors influencing attrition in that department.</p><p>- Employee-Level Data: A table provides a view of top performers, showing their Performance Rating and Customer Satisfaction scores.</p><p>The link to the interactive dashboard <a href="https://app.powerbi.com/view?r=eyJrIjoiNjIwZTljMDQtZjEyOC00NjA4LWJkYmMtMjVkMTE2NWQ3OGRlIiwidCI6IjQ2NTRiNmYxLTBlNDctNDU3OS1hOGExLTAyZmU5ZDk0M2M3YiIsImMiOjl9">here</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*J-u4I0FFZZGuAMmPvAJokA@2x.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rtoTOKbPkb5Vkd40kYLPIA@2x.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jmbmm8_8lxiMKKXNk1P13Q@2x.jpeg" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=733b4bed2cc5" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[I was interviewed by aspiring data analysts and here is what i had to say- this might help you!]]></title>
            <link>https://medium.com/@odunthedatagirl/i-was-interviewed-by-aspiring-data-analysts-and-here-is-what-i-had-to-say-this-might-help-you-c2a29a399907?source=rss-42bca778fa7e------2</link>
            <guid isPermaLink="false">https://medium.com/p/c2a29a399907</guid>
            <dc:creator><![CDATA[Odunayo Aliu]]></dc:creator>
            <pubDate>Wed, 21 May 2025 03:25:10 GMT</pubDate>
            <atom:updated>2025-05-21T03:25:10.224Z</atom:updated>
            <content:encoded><![CDATA[<ol><li>My Journey into Data Analysis started with my love for mathematics and solving problems using numbers, my roommate was having a discussion with another roommate and i heard about how data analysis is related to mathematics in some way. Immediately it got my attention. I started googling about data analysis and data science. Started learning python on scrimba and udemythen took a step forward to applying for scholarships on coursera and Datacamp. I got the scholarships and learned at a very high pace. Data Analysis was very interesting to me (it still is). I took my forst certification exam on datacamp and i failed at my first trial, on the practical aspects. Didn’t give up, i retook the exam and passed it the second time. I always posted my learning progress on my twitter account and linkedin, work on real life projects like covid 19 as it was. a worldwide pandemic at the time. This helped me so much that when I finally got certified I got a job, online presence and my growing portfolio.</li><li>Python and Power Bi are my best tools but the jobs i have gotten even till date focuses on google sheets and excel. So i started teaching via my YouTube channel and a few mentorship. It has been quite a journey actually(if i must say)</li><li>2.	Career Opportunities in Data Analysis Today</li></ol><p>From my perspective, data analysis is wild spreading career that every organization needs one, both here in Nigeria and globally. Data has become the lifeblood of modern organizations, and the demand for professionals who can extract meaningful insights from this data is only increasing.</p><p>In Nigeria: We’re seeing a growing awareness of the power of data across various sectors, from fintech and e-commerce to agriculture and healthcare. Companies are starting to invest more in building data analytics capabilities, leading to an increasing number of entry-level and mid-level roles. The rise of tech hubs and initiatives focused on digital transformation is also fueling this growth. While the market might still be maturing compared to more established economies, the potential is immense, and early career data analysts have a fantastic opportunity to shape the future of data-driven decision-making here.</p><p>Globally: The demand for data analysts is well-established and continues to surge. Organizations of all sizes, across every industry, are recognizing the strategic advantage that data insights can provide. This translates to a wide range of opportunities, from specialized roles in areas like marketing analytics or financial analysis to more general data analyst positions. Furthermore, the increasing sophistication of data science and machine learning creates a strong pipeline for data analysts to upskill and advance their careers. The remote work trend has also opened up global opportunities for analysts who are proficient in the required tools and have strong communication skills.</p><p>However, it’s also important to note that the field is becoming more competitive. Employers are looking for candidates with a strong blend of technical skills (SQL, Python, visualization), analytical thinking, and the ability to communicate complex findings clearly. This means continuous learning and upskilling are crucial for staying relevant.</p><p>3.	The Most Common Trait of Successful Data Analyst.</p><p>If I had to pinpoint one common trait among the most successful data analysts I’ve encountered, it would be intellectual curiosity and willingness to learn.</p><p>You have to be able to think out of the box, learn, relearn and unlearn. Learning has to do with curiosity, regardless of how much you learn there is always something you don’t know yet. Also boldness (or i’d call it audacity)</p><p>Being audacious has a lot to do with knowing how to pitch yourself to people around you.</p><p>4.	My Practical Advice for Aspiring Data Analysts.</p><p>If you’re just starting out as a data analyst, take a step at a time, don’t skip any process; start with the core concepts: excel, SQL, and a programming language like Python and a visualization tool like power bi/ tableau. Another advice I’d give is consistency in practice especially with real world datasets (because they are usually what helps you in the real world). Work on personal projects using publicly available datasets. Platforms like Kaggle, Google Dataset Search, and even government open data portals offer a wealth of data to explore.</p><p>Building a Portfolio: Your portfolio is your proof of work. Showcase your projects on platforms like GitHub even on every social media platform. Also networking which involves connecting with other learners and professionals on platforms like LinkedIn, Twitter, and data science forums. Ask questions, share your learnings, and participate in discussions. The support and insights you gain from the community are invaluable. It’s also important that you document your progress,just like I did, consider documenting your learning journey through blog posts or social media. This helps solidify your understanding and can also attract opportunities.</p><p>Focus on Communication: Data analysis isn’t just about numbers; it’s about telling a story with data. Practice communicating your findings clearly and concisely to both technical and non-technical audiences. Visualization skills are crucial here.</p><p>Be Patient and Persistent: Learning data analysis takes time and effort. Don’t get discouraged by initial challenges. Consistency and a growth mindset are essential.</p><p>Seek Out Mentorship (If Possible): Having someone with more experience guide you can be incredibly beneficial. Look for mentors within the data community.</p><p>Stay Curious: Never stop learning! The field of data is constantly evolving, so embrace a mindset of continuous learning. Follow industry blogs, take new courses, and experiment with new tools.</p><p>5.	My Personal Professional Aspirations</p><p>Looking ahead, my personal professional aspirations are centered around growth and impact. I currently have a YouTube channel which i look forward to building a strong community of data analysts where i get to help them ace their growth and learning process with my tutorial videos and tips.</p><p>Ultimately, my goal in any organization is to be in a position where I can significantly influence strategic decisions within an organization through insightful data analysis. I want to help businesses leverage their data to achieve their goals and create positive impact.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c2a29a399907" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Inevitable Growth: How to become a power BI professional in one year]]></title>
            <link>https://medium.com/@odunthedatagirl/inevitable-growth-how-to-become-a-power-bi-professional-in-one-year-46ffabe81be6?source=rss-42bca778fa7e------2</link>
            <guid isPermaLink="false">https://medium.com/p/46ffabe81be6</guid>
            <dc:creator><![CDATA[Odunayo Aliu]]></dc:creator>
            <pubDate>Fri, 27 Sep 2024 08:36:47 GMT</pubDate>
            <atom:updated>2024-09-27T08:36:47.675Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Inevitable Growth: How to become a power BI professional in one yea</strong>r</p><p>After lots of practice, I tried to recreate a dashboard from July 2023 and discovered how much I have grown in the past year. I moved from this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qHNbGEnJO_GxFk3D8IQ90w@2x.jpeg" /></figure><p>A powerBI dashboard for my first portfolio project.</p><p>I tried to recreate the dashboard for my YouTube tutorial video and I realized how much I have improved with data visualization.</p><p>I effortlessly recreated this and you don’t have to be a stakeholder to spot the differences</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*56XW6LsHsRJibqw8OgtnHQ@2x.jpeg" /></figure><p>The labels, colour arrangements and effective representation of visuals are very important when designing a dashboard.</p><p>Constant practice helps growth and you can only imagine how much you have grown when you compare.</p><p>The full tutorial video on this dashboard is live on my YouTube channel: <a href="https://youtube.com/@odunthedatagirl?si=KAHUuDKZ_XiE9Phq">https://youtube.com/@odunthedatagirl?si=KAHUuDKZ_XiE9Phq</a></p><p>Thank you for reading!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=46ffabe81be6" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Exploratory Data Analysis using Python: A step by step Vehicle Sales Project]]></title>
            <link>https://medium.com/@odunthedatagirl/exploratory-data-analysis-using-python-a-step-by-step-vehicle-sales-project-e98a812269f0?source=rss-42bca778fa7e------2</link>
            <guid isPermaLink="false">https://medium.com/p/e98a812269f0</guid>
            <category><![CDATA[data-analysis-project]]></category>
            <category><![CDATA[exploratory-data-analysis]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[power-bi]]></category>
            <category><![CDATA[python-programming]]></category>
            <dc:creator><![CDATA[Odunayo Aliu]]></dc:creator>
            <pubDate>Mon, 29 Apr 2024 02:46:05 GMT</pubDate>
            <atom:updated>2024-05-22T12:23:35.439Z</atom:updated>
            <content:encoded><![CDATA[<h3>Exploratory Data Analysis using Python: A step-by-step process of a Vehicle Sales Project</h3><p>Exploratory Data Analysis is an important aspect of data analysis that involves statistical summarization and examination of what the database expresses. There are various tools available for exploratory data analysis, but I believe Python makes it quick, simple and easy.</p><p>First approach was setting up Jupyter Notebook (I used Anaconda Navigator), and then uploading the document to the home page. I created a ipynb page (a notebook) and imported the database.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*oJBVx780nU8Guvsty4sChg@2x.jpeg" /></figure><p>2. After importing, I discovered that there were 26 columns instead of 19, so that makes it 7 null columns, the best solution to that problem is dropping the columns, I dropped them then I used the shape function to reaffirm the columns I have</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/417/1*gGRrkDB1fOn2ct8Gfg1HCw@2x.jpeg" /></figure><p>3. I checked for the statistical summary and null values in the database to ensure accuracy with further analysis.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/417/1*jvk1_mkIqnhbQzuRe3XRmQ@2x.jpeg" /></figure><p>4. I confirmed the value counts for text-based columns to confirm spelling errors.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/417/1*SFYuY2tHTbDn2fg5k_suKQ@2x.jpeg" /></figure><p>Summary: Exploratory Data Analysis helps to easily identify the problems your data will be solving, the questions to be asked and the possible solutions and insights to be adopted from the data.</p><p>I designed an interactive dashboard with power BI to answer a few questions about the sales</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/540/1*C6N5Y4YYl1fNYV7hBwmOIg@2x.jpeg" /></figure><p>Thank you for reading, I hope this is helpful for project ideas on exploratory data analysis.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e98a812269f0" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[10 Power BI Interactive Dashboards for Portfolio Project]]></title>
            <link>https://medium.com/@odunthedatagirl/10-power-bi-interactive-dashboards-for-portfolio-project-534eafa23daf?source=rss-42bca778fa7e------2</link>
            <guid isPermaLink="false">https://medium.com/p/534eafa23daf</guid>
            <category><![CDATA[power-bi-dashboard]]></category>
            <category><![CDATA[power-bi-tutorials]]></category>
            <category><![CDATA[power-bi]]></category>
            <category><![CDATA[power-bi-data-analyst]]></category>
            <category><![CDATA[power-bi-project]]></category>
            <dc:creator><![CDATA[Odunayo Aliu]]></dc:creator>
            <pubDate>Fri, 19 Apr 2024 19:59:10 GMT</pubDate>
            <atom:updated>2024-04-29T02:51:15.732Z</atom:updated>
            <content:encoded><![CDATA[<p>I selected ten Power BI dashboards I designed; I obtained my datasets from Kaggle. You can use them as an inspiration for your portfolio project.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nLGezfbveD_Bm3FW2DsvdQ.jpeg" /><figcaption>Students Performance Dashboard</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Pbyzr3vp3ei8juDE0Nmm1Q.png" /><figcaption>Airways Dashboard</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*33Cw_T2h5PBXHVF1hHYtzw.jpeg" /><figcaption>Valentine’s Day Dashboard</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*C9MtXmiCAe4qwfDBDDZeGw.jpeg" /><figcaption>Wastage Analysis</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/1*AxKZrZiT-QayAlKHIj7HzQ@2x.jpeg" /><figcaption>Financial Analysis</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/540/1*C6N5Y4YYl1fNYV7hBwmOIg@2x.jpeg" /><figcaption>Vehicle Sales Dashboard</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*kMUPbCPJX-Aj3Xxme0eYcA.jpeg" /><figcaption>Sales Dashboard</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*o3tFTtXkKieCMCHY3jINnw.jpeg" /><figcaption>Salary Analysis</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/319/1*4ZWp7fxAlrbFSo9pTRHNVA@2x.jpeg" /><figcaption>Car Sales Dashboard</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/320/1*0S0OORK55E7JReE5wFY3CQ@2x.jpeg" /><figcaption>Menstruation Analysis</figcaption></figure><p>Thank you!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=534eafa23daf" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[10 Data Visualization Mistakes Data Analysts should take note of]]></title>
            <link>https://medium.com/@odunthedatagirl/10-data-visualization-mistakes-data-analysts-should-take-note-of-bdb215d3e6e0?source=rss-42bca778fa7e------2</link>
            <guid isPermaLink="false">https://medium.com/p/bdb215d3e6e0</guid>
            <category><![CDATA[data-analyst]]></category>
            <category><![CDATA[data-visualization-tips]]></category>
            <category><![CDATA[data-visualization]]></category>
            <dc:creator><![CDATA[Odunayo Aliu]]></dc:creator>
            <pubDate>Mon, 04 Mar 2024 21:47:11 GMT</pubDate>
            <atom:updated>2024-03-05T20:54:31.310Z</atom:updated>
            <content:encoded><![CDATA[<p>Data Visualization is one of the most important part of data analysis and it is important to pay close attention to the possible mistakes that are most likely to be made while creating visualizations.</p><p><strong>1. Choosing the wrong chart type:</strong> Different chart types are better suited for different types of data and purposes. Using a pie chart for data that shows trends over time, for example, can be misleading and should be avoided.</p><p><strong>2. Misleading colour choices:</strong> Colours can evoke emotions and influence how viewers interpret data. Using colours that are too similar or have cultural connotations that could be misinterpreted is a mistake.</p><p><strong>3. Cluttering the visualization:</strong> Including too much information, using unnecessary visual elements like excessive decorations, or having poor labelling can overwhelm viewers and make it difficult to understand the key message.</p><p><strong>4. Distorting data:</strong> Manipulating data scales, axes, or truncating data points can misrepresent the actual findings and mislead viewers.</p><p><strong>5. Omitting context:</strong> Data visualizations need context to be fully understood. This includes providing clear titles, labels, legends, and any necessary background information about the data and its source.</p><p><strong>6. Ignoring accessibility:</strong> Not considering colour blindness or other visual impairments when choosing colours and designing the visualization can exclude certain audiences from understanding the information.</p><p><strong>7. Using 3D charts unnecessarily:</strong> While 3D charts can be visually appealing, they can also distort the data and make it difficult to compare values accurately. In most cases, simpler 2D charts are more effective.</p><p><strong>8. Not considering the audience:</strong> It’s crucial to tailor your data visualization to your target audience. Consider their level of familiarity with data and the complexity they can handle to ensure clear communication.</p><p><strong>9. Overusing data visualizations:</strong> Not every piece of information needs to be presented visually. Sometimes, a simple table or text description may be more appropriate and effective.</p><p><strong>10. Not iterating and testing:</strong> The first draft of your data visualization is unlikely to be perfect. Be prepared to iterate on your design based on feedback and test it with your target audience to ensure it is clear and effective.</p><p>Conclusion: Data Visualization is an engaging process hence the visuals should be clear, concise, and highly interactive. A dashboard for instance should have more visuals than text to captivate our stakeholders and viewers.</p><p>Thanks for reading.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bdb215d3e6e0" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[PCOS Prediction / Menstruation Analysis]]></title>
            <link>https://medium.com/@odunthedatagirl/pcos-prediction-menstruation-analysis-6f1cf773c1a3?source=rss-42bca778fa7e------2</link>
            <guid isPermaLink="false">https://medium.com/p/6f1cf773c1a3</guid>
            <category><![CDATA[data-analysis]]></category>
            <category><![CDATA[pcos]]></category>
            <category><![CDATA[menstruation]]></category>
            <category><![CDATA[menstrual-health]]></category>
            <category><![CDATA[pcos-awareness]]></category>
            <dc:creator><![CDATA[Odunayo Aliu]]></dc:creator>
            <pubDate>Wed, 21 Feb 2024 21:29:33 GMT</pubDate>
            <atom:updated>2024-02-22T10:41:02.020Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*MiKGO_bjPDofHEXqCN3dfA.jpeg" /></figure><p>The rapid increase in PCOS disorders among young adults around the world has led to this analysis. Its purpose is to support women’s reproductive health by examining menstrual patterns and identifying factors that contribute to hormonal imbalances and PCOS risk.</p><p>It appears that most people in this dataset have regular menstrual cycles without experiencing hormonal imbalances. Only about 2% of individuals report unusual or irregular menstrual flow. The age range of the participants is between 14 and 23, which could be a relevant factor.</p><p>Based on this analysis, the following factors may impact the menstrual cycle:</p><p>1. Body Mass Index (BMI): BMI is a measure of body weight in kilograms divided by the square of height in meters. Individuals with a high BMI tend to have high body fatness. Recent studies suggest that underweight individuals have longer cycle lengths compared to those in the normal weight range and overweight individuals. This difference could be attributed to differences in blood flow among individuals of different weight categories.</p><p>2. Age: Age is a crucial factor in hormonal imbalances. The analysis shows that older women have higher luteal phase and lower estimated ovulation days.</p><p>3. Period Length/Cycle Length: If the period length is more than 5 days, it could indicate an abnormality. It is essential to consider eating habits, blood counts, and seek medical attention to prevent the risk of PCOS.</p><p>Conclusion: There are several other factors that affect menstrual flow and the risk of hormonal imbalances among young women. It is important to pay attention to the luteal days and estimated days to ovulation to predict abnormalities and menstruation problems.</p><p>You can access the Interactive Dashboard through this link: <a href="https://app.powerbi.com/groups/me/reports/927ffc43-fa38-4a27-8ded-d81c51198d9a/ReportSection?experience=power-bi">here</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6f1cf773c1a3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Covid-19 Data Analysis Project using Excel, Python, and Power BI]]></title>
            <link>https://medium.com/@odunthedatagirl/covid-19-data-analysis-project-using-excel-python-and-power-bi-bc9a6b3cf656?source=rss-42bca778fa7e------2</link>
            <guid isPermaLink="false">https://medium.com/p/bc9a6b3cf656</guid>
            <category><![CDATA[data-analysis-portfolio]]></category>
            <category><![CDATA[data-analysis]]></category>
            <category><![CDATA[data-analysis-with-python]]></category>
            <category><![CDATA[covid-19-analysis]]></category>
            <category><![CDATA[data-analysis-excel]]></category>
            <dc:creator><![CDATA[Odunayo Aliu]]></dc:creator>
            <pubDate>Sun, 30 Jul 2023 20:08:55 GMT</pubDate>
            <atom:updated>2024-05-24T13:31:43.314Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/220/1*0pRttVxb426fpiDdZxt0Gg.jpeg" /></figure><p>COVID-19 is a highly infectious respiratory illness caused by the novel coronavirus. It first emerged in Wuhan, China, in December 2019 and has since spread worldwide. The symptoms of COVID-19 include fever, cough, and difficulty breathing. It’s important to practice social distancing, wear masks, and follow health guidelines to help stop the spread of the virus. COVID-19 spread rapidly across all continents in the world, The confirmed cases and deaths varied across different countries depending on population, different weather conditions, and adherence to safety guidelines.</p><p>This Analysis was made on the 21st of July 2023. Although, the number of confirmed cases and deaths is still in progression across the world. The purpose of this analysis is to develop insights into the confirmed cases, The deaths trend over the years and the confirmed cases Versus deaths according to the population across the countries in the world from January 2020 till July 2023.</p><p>Approach: The Questions to be asked based on this analysis are:</p><ol><li>What was the total number of confirmed cases and deaths in the world?</li><li>Which countries had the highest number of confirmed cases and Deaths across the world?</li><li>At what rate did COVID-19 spread across all the continents in the world?</li><li>How did Covid-19 affect countries based on population?</li></ol><p>5. How did Covid-19 spread trend from 2020 to 2023?</p><h3>Process Methodology</h3><p>I used the following steps to carry out the analysis:</p><p>. Data Collection</p><p>. Data Cleaning</p><p>. Data Validation and Exploration</p><p>. Visualization and Dashboard Creation</p><p>. Data Collection: The data collection process was from the internet, ourworldindata.org</p><p>. Data Cleaning: The tool used: Excel. The cleaning process involved deleting rows with null values from the continent column, deleting columns that were not needed for this analysis and converting data types.</p><p>. Data Validation and Exploration: I used Python for Data Exploration</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/950/1*SB3UbaHceZwLejxMBDv4_A.png" /></figure><p>The dataset was imported into Jupyter Notebook to cross-check if there are any inconsistent values or irrelevant columns.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/940/1*7QEcPGwPm7AMtRpvxSDsOw.png" /></figure><p>I checked if there were any unexpected values in the continents column and countries column(location), and they were all valid.</p><p>I also checked for null values in the dataset, the total_vaccinations column had the highest number of null values because most of the countries didn’t keep records of the number of vaccinations that were used to fight the disease.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/934/1*HU0r7et0aF4rGPam7bJxCQ.png" /></figure><p>Visualization and Dashboard Creation:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/984/1*W86kXarTZTA63kz3vHxswg@2x.jpeg" /></figure><p>The tool used for this stage is Power BI. The dashboard covers the answers to the questions asked earlier.</p><p>From the dashboard, we got insights and answers to our questions with this data-driven dashboard.</p><ol><li>The total confirmed cases were 770,432,109 while the total deaths were 7,079,107 which shows that 0.91% of the people that covid-19 died from the disease.</li><li>United States, China, India, France and Brazil had the highest number of confirmed cases among the 243 countries that got affected by covid 19.</li><li>Covid-19 spread most in Asian Countries with 300 million confirmed Cases and 1.8 million deaths, Europe had a total of 249 million total confirmed cases and 2.1 million deaths, North America had a total of 124 million total cases and 1.4 million deaths, South America had a total of 68.8 million confirmed cases and 1.6 million deaths, Oceania had a total of 14 million total cases and 28.7 million deaths while Africa had the lowest number of total confirmed cases with 13 million total cases and 259 thousand deaths.</li><li>Covid-19 affected China, India and United States more than other countries based on the origin of the disease and population.</li><li>The trend went high from 2020 to 2021 and dropped drastically from Q3 in 2021 to 2023, This was a result of vaccination usage and proper safety practices all over the world.</li></ol><p>The link to the responsive dashboard:<a href="https://app.powerbi.com/groups/me/reports/ecf5a01d-2921-45af-bd2e-ae3e0a9e62f8/ReportSection08b0ae75897ed5d23091?experience=power-bi"> covid — Power BI</a></p><p>Thanks for reading.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bc9a6b3cf656" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>