<?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 Spandana Tej Rameswaram on Medium]]></title>
        <description><![CDATA[Stories by Spandana Tej Rameswaram on Medium]]></description>
        <link>https://medium.com/@spandanatej.rameswaram?source=rss-60448ba6ec1c------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*FGVj3RLjxBsp10HfVYDoWQ@2x.jpeg</url>
            <title>Stories by Spandana Tej Rameswaram on Medium</title>
            <link>https://medium.com/@spandanatej.rameswaram?source=rss-60448ba6ec1c------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 07:35:56 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@spandanatej.rameswaram/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[Mastering Data Cleaning With Pandas]]></title>
            <link>https://medium.com/@spandanatej.rameswaram/mastering-data-cleaning-with-pandas-c38bb99d93d8?source=rss-60448ba6ec1c------2</link>
            <guid isPermaLink="false">https://medium.com/p/c38bb99d93d8</guid>
            <dc:creator><![CDATA[Spandana Tej Rameswaram]]></dc:creator>
            <pubDate>Wed, 21 Feb 2024 00:57:12 GMT</pubDate>
            <atom:updated>2024-02-21T00:57:12.854Z</atom:updated>
            <content:encoded><![CDATA[<p>Cleaning data is an essential stage in the process of analyzing data, encompassing the identification and rectification of errors, disparities, and absent values within the dataset. Pandas, an influential Python library, presents an extensive array of tools and features for effectively tidying and preprocessing data.</p><p>Pandas is a powerful and popular open-source library for data manipulation and analysis in Python. It’s built on top of NumPy and provides easy-to-use data structures and functions that are essential for working with structured data.</p><p>In this Blog, we’ll delve into diverse methods and optimal strategies for data cleaning with Pandas, equipping to prime the data for perceptive analysis and modeling endeavors.</p><p><strong>1.Loading Data:</strong></p><p>The first step in data cleaning is loading the dataset into a Pandas DataFrame. But, before that, Pandas Library has to be imported. Pandas provides functions like read_csv(), read_excel(), and read_sql() to read data from different file formats and data sources.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mDzbO4-3bnInMUm9ZK8P8A.png" /><figcaption>Importing pandas and loading data</figcaption></figure><p><strong>2. Exploring the Data:</strong></p><p>As the data is loaded, Now we can explore the dataframe to understand its structure, features, and any potential issues. Pandas offers methods like head(), info(), and describe() for this purpose.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0hElHbY949GrY-T9tEFp8w.png" /><figcaption>Exploring the data</figcaption></figure><p><strong>3. Handling Missing Data:</strong></p><p>Finding and addressing the missing the values in the dataframe is one of the important steps to be followed before the analysis. Pandas provides methods like isnull(), notnull(), dropna(), and fillna() to handle missing data effectively.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*AnsAe_vy_bdfYkzyLyOL4w.png" /><figcaption>Handling missing Data</figcaption></figure><p><strong>4. Handling Duplicates:</strong></p><p>Another important step in Data cleaning for better analysis is Handing the duplicates effectively.Pandas offers the drop_duplicates() method to remove duplicate rows from the DataFrame.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uwSHYX6CO_80FOROptGfjA.png" /><figcaption>Handling Duplicates</figcaption></figure><p><strong>5. Data Transformation:</strong></p><p>Data transformation involves converting data types, renaming columns, and splitting or combining columns. Pandas provides methods like astype() for converting the datatypes, rename() for renaming the columns, and string manipulation functions for splitting or combining the columns.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*LUS4M9Ru3XNCvmjVZ2uk1Q.png" /><figcaption>Examples for Data Transformation</figcaption></figure><p><strong>6. Handling Outliers:</strong></p><p>Outliers can explicitly affect the analysis if not handled well. Pandas offers methods like quantile() to detect the outliers base on the requirement and the data, and clip() to handle the outliers in the dataset.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7NjnLMAqPv0W_2DBsST3kw.png" /><figcaption>Outliers Explanation with Examples</figcaption></figure><p><strong>7. Data Normalization and Standardization:</strong></p><p>Data normalization and standardization are preprocessing techniques used to re-scale the values of numerical features in a dataset, making them suitable for machine learning algorithms and statistical analysis. While both techniques aim to transform the data into a common scale, they differ in their approach and the resulting distributions, where Normalization re-scales the values of a feature to a range between 0 and 1 and where as Standardization transforms the values of a feature to have a mean of 0 and a standard deviation of 1. It is beneficial when the features have different scales and follow a Gaussian distribution (bell curve). Pandas enables these operations using arithmetic operations.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*72qq4Bv_wVH77ul1lk3YSQ.png" /><figcaption>Normalization and Standardization of Data</figcaption></figure><p><strong>8. Exporting Cleaned Data:</strong></p><p>Once the data cleaning process is complete, the cleaned data can be exported to a new file using methods like to_csv(), to_excel(), etc.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DGCLk6tbUzFXG6Fph15obQ.png" /><figcaption>Exporting Cleaned Data</figcaption></figure><p><strong>Conclusion:</strong></p><p>Data cleaning stands as an essential stride in the journey of data analysis, and Pandas furnishes an extensive arsenal to execute this task with precision. Harnessing the robust functions and methods within Pandas, can seamlessly tackle prevalent data discrepancies, uphold data integrity, and prime the dataset for insightful analysis and revelations. Armed with the insights from this blog, we can confront data cleaning hurdles adeptly and unlock the boundless possibilities within the data analysis endeavors.</p><p>Happy Cleaning &amp; Happy Learning!!</p><p><strong>References:</strong></p><p><a href="https://pandas.pydata.org/docs/">pandas documentation — pandas 2.2.0 documentation (pydata.org)</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c38bb99d93d8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Harnessing the Potential of Data Blending in Tableau]]></title>
            <link>https://medium.com/@spandanatej.rameswaram/harnessing-the-potential-of-data-blending-in-tableau-e2e2c2bdb449?source=rss-60448ba6ec1c------2</link>
            <guid isPermaLink="false">https://medium.com/p/e2e2c2bdb449</guid>
            <category><![CDATA[tableau]]></category>
            <category><![CDATA[data-analyst]]></category>
            <category><![CDATA[data-blending]]></category>
            <dc:creator><![CDATA[Spandana Tej Rameswaram]]></dc:creator>
            <pubDate>Fri, 16 Feb 2024 19:54:57 GMT</pubDate>
            <atom:updated>2024-02-16T19:54:57.710Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2lIHY4Rqx9jYEEUwBVSzNw.jpeg" /></figure><p>Data blending in Tableau is a powerful technique that allows to combine data from different sources seamlessly, enabling deeper insights and more comprehensive analysis. Whether it is with disparate databases, Excel spreadsheets, or cloud-based applications, data blending empowers to unlock hidden correlations and trends. In this blog, we’ll delve into the intricacies of data blending in Tableau, exploring its benefits, best practices, and practical applications.</p><p>The simplest way to think of data blending is as an aggregated outer join. However, rather than creating a physical join of the data set, a data blend is a view of an aggregated join from disparate data sets. Data blending is particularly useful when the blend relationship — the linking fields — need to vary on a sheet-by-sheet basis, or when combining already published data sources.</p><p><strong><em>Understanding Data Blending:</em></strong></p><p>Data blending refers to the process of combining data from multiple sources within Tableau to create a unified view for analysis. Unlike the traditional joins, which require a common field to merge datasets, data blending allows users to link information based on related dimensions. This flexibility is particularly valuable when working with datasets that lack a common identifier or when integrating data from diverse sources.</p><p>Blends query each data source independently, aggregate the results to the appropriate level, then present the results together visually in the view. Because of this, blends can handle different levels of detail and also work with published data sources. Blends don’t create a new, blended data source (and therefore can’t be published as a “blended data source”). Instead, they are simply blended results visualized per sheet.</p><p><strong><em>Steps for Data Blending:</em></strong></p><p>Data blending is executed on a per-sheet basis and is initiated when a field from a secondary data source is utilized in the visualization.</p><p>Here’s the step-by-step process involved in Data Blending,</p><p><strong>1. Connect to Data Sources:</strong></p><ul><li>Launch Tableau and navigate to the “Data” pane.</li><li>Data can be added by following this — Data &gt; New data source.</li><li>Repeat the process to connect to additional data sources that has to be blended.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NI2jp7ay33tbEsOI5xF26A.png" /></figure><p><strong>2. Define Primary and Secondary Data Sources:</strong></p><ul><li>Drag a field from your primary data source onto the worksheet. This action automatically designates it as the primary data source.</li><li>Switch to the secondary data source by clicking on its tab at the bottom of the data pane.</li><li>The primary data source is indicated with a blue check mark on the data source. Fields from the primary data source used in the view have no indication.</li><li>The secondary data source is indicated with an orange check mark on the data source and an orange bar down the side of the Data pane. Fields from the secondary data source used in the view have an orange check mark.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9T52YPtHHcBSiR6Hg9cHqw.png" /></figure><p><strong>3. Add Secondary Data to the View:</strong></p><ul><li>From the secondary data source, drag a field onto the worksheet to include it in the visualization. This action designates the second data source as the secondary source.</li><li>As on instance, if the primary data source has a “Day” field that only contains Monday, Wednesday, and Friday, any view built around Days will only display these three (Monday, Wednesday, and Friday), even if the secondary data source has values for 7 Days of the week. If the desired analysis involves all 7 Days, better practice is to switch data source with 7 Days of the week as the primary Data Source.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*aJLbz4P2fwvlJiIDMFfppg.png" /></figure><p><strong>4. Identify Common Fields for Blending:</strong></p><ul><li>Look for fields that are common between the primary and secondary data sources. These common fields will be used to blend the data.</li><li>If there is a linking field icon (), the data sources are automatically linked. As long as there is at least one active link, the data can be blended.</li><li>If there are broken link icons (), click the icon next to the field that should link the two data sources. The slash will go away, representing an active link.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/749/1*20lHE559oy6zaX4EC1u1YA.png" /></figure><p><strong>5. Create Blended Calculations:</strong></p><ul><li>To use fields from both data sources in calculations or visualizations, create blended calculations. These calculations combine data from both sources.</li><li>Right-click on a field in the data pane and select “Create Calculated Field” to create a blended calculation.</li><li>The Drop-down button in the Data pane is also used to create the required Calculated Field.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/763/1*vDxcnq1Q0Tul51jJAlSCwA.png" /></figure><p><strong>6. Customize Blend Relationships:</strong></p><ul><li>You can customize blend relationships by clicking on the linking icon that appears next to blended fields in the data pane. This allows you to specify how the data should be blended based on the relationship between fields.</li></ul><p><strong>7. Build Visualizations:</strong></p><ul><li>Use the blended data fields to create visualizations just the way it can be done with a single data source.</li><li>Drag and drop fields onto shelves, choose chart types, apply filters, and customize the visualization as desired.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gVnxFRLW7r43HmcieHf6Jw.png" /></figure><p><strong>8. Review and Validate:</strong></p><ul><li>Review the blended data visualization to ensure that it accurately represents the insights required.</li><li>Validate the blended data by comparing it with the original datasets and confirming that the blended results align with expectations.</li></ul><p><strong>9. Save and Share:</strong></p><ul><li>After the best visualization is created, the workbook can be saved in Tableau format (.twb or .twbx) and shared with others for further analysis or to meet the purpose of visualization.</li></ul><p><strong><em>Benefits of Data Blending:</em></strong></p><p><strong>1.Flexibility:</strong></p><p>Data blending accommodates disparate datasets, enabling users to integrate information without restructuring the original sources.</p><p><strong>2. Deeper Insights:</strong></p><p>By blending data from different sources, analysts can uncover correlations and patterns that may not be apparent when examining individual datasets.</p><p><strong>3. Efficiency:</strong></p><p>Data blending streamlines the analysis process, eliminating the need for complex data preprocessing or manual data manipulation.</p><p><strong>4. Dynamic Analysis:</strong></p><p>Tableau’s interactive features allow users to explore blended data dynamically, facilitating iterative analysis and hypothesis testing.</p><p><strong>5. Scalability:</strong></p><p>Whether you’re working with small-scale datasets or enterprise-level data sources, Tableau’s data blending capabilities scale to meet your analytical needs.</p><p><strong><em>Best Practices for Data Blending:</em></strong></p><p>Following the below steps would be helpful in the process of Blending the Data from different sources.</p><p><strong>1.Understand the Data:</strong></p><p>Before embarking on data blending endeavors, it’s paramount to comprehensively grasp the underlying data structures and interrelationships. A clear understanding of the data’s composition aids in identifying commonalities and potential challenges during the blending process.</p><p><strong>2. Identify Key Dimensions:</strong></p><p>Successful data blending hinges upon the identification of key dimensions or fields that serve as the linchpin for amalgamating disparate datasets. By pinpointing commonalities across datasets, analysts can effectively align and merge relevant data points.</p><p><strong>3. Maintain Data Integrity:</strong></p><p>Preserving data integrity stands as a fundamental pillar throughout the blending process. Rigorous validation procedures should be implemented to verify the accuracy, consistency, and reliability of blended datasets. By ensuring data fidelity, analysts can foster trust in the insights derived from blended data.</p><p><strong>4. Optimize Performance:</strong></p><p>Streamlining performance constitutes a pivotal aspect of data blending endeavors. To enhance efficiency, it’s advisable to curate blended views by minimizing the inclusion of extraneous records and fields. Leveraging Tableau’s data extract capabilities proves beneficial, particularly when working with voluminous datasets, thereby optimizing performance and responsiveness.</p><p><strong>5. Test and Iterate:</strong></p><p>Embracing an iterative approach facilitates the refinement and enhancement of blended data analyses. Analysts should actively experiment with diverse blending techniques and configurations to discern optimal strategies for deriving actionable insights. Through continuous testing and iteration, analysts can iteratively refine their blending methodologies, leveraging newfound insights to propel data-driven decision-making.</p><p><strong><em>Practical Applications of Data Blending:</em></strong></p><p><strong>1. Sales and Marketing Analysis:</strong></p><p>Blend sales data with marketing campaign metrics to analyze the impact of promotional activities on revenue.</p><p><strong>2. Customer Segmentation:</strong></p><p>Combine demographic data with purchasing behavior to identify distinct customer segments and tailor marketing strategies accordingly.</p><p><strong>3. Supply Chain Optimization:</strong></p><p>Integrate inventory data with shipping and logistics information to optimize supply chain operations and minimize costs.</p><p><strong>4. Financial Reporting:</strong></p><p>Blend financial data from multiple sources to create comprehensive reports and dashboards for stakeholders.</p><p><strong>5. Healthcare Analytics:</strong></p><p>Combine patient demographics with medical records to identify trends and patterns in disease prevalence and treatment outcomes.</p><p><strong><em>Conclusion:</em></strong></p><p>Data Blending in Tableau presents a robust framework tailored to seamlessly merge diverse datasets, thereby unlocking profound insights. Empowered by Tableau’s user-friendly interface and robust analytical features, individuals can seamlessly integrate data from various sources. This amalgamation empowers informed decision-making and unveils concealed opportunities. Armed with adept knowledge of best practices and real-world applications, analysts can fully exploit the potential of data blending, thereby gaining a competitive advantage in today’s data centrist environment.</p><p><strong><em>References:</em></strong></p><p><a href="https://help.tableau.com/current/pro/desktop/en-us/multiple_connections.htm">https://help.tableau.com/current/pro/desktop/en-us/multiple_connections.htm</a></p><p><a href="https://www.tableau.com/drive/intro-data-blending">https://www.tableau.com/drive/intro-data-blending</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e2e2c2bdb449" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>