<?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 Statsvenu on Medium]]></title>
        <description><![CDATA[Stories by Statsvenu on Medium]]></description>
        <link>https://medium.com/@venugopal-manneni?source=rss-8f0d88203c4c------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/2*ScTMeCIeItdWRhanyZECjg.jpeg</url>
            <title>Stories by Statsvenu on Medium</title>
            <link>https://medium.com/@venugopal-manneni?source=rss-8f0d88203c4c------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 03:51:02 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@venugopal-manneni/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[The Evolution of Analytics: From Hard-Coded Scripts to Conversational AI]]></title>
            <link>https://venugopal-manneni.medium.com/the-evolution-of-analytics-from-hard-coded-scripts-to-conversational-ai-d36a7cc8a07a?source=rss-8f0d88203c4c------2</link>
            <guid isPermaLink="false">https://medium.com/p/d36a7cc8a07a</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[statistical-analysis]]></category>
            <category><![CDATA[trends]]></category>
            <dc:creator><![CDATA[Statsvenu]]></dc:creator>
            <pubDate>Sat, 22 Mar 2025 09:24:04 GMT</pubDate>
            <atom:updated>2025-03-22T09:24:04.467Z</atom:updated>
            <content:encoded><![CDATA[<p>In the ever-evolving landscape of analytics, the way we conduct data analysis has transformed dramatically over the years. As technology progresses, so does the way we interact with analytical tools, making them more accessible, efficient, and intelligent. This transformation can be categorized into three major generations of analytics, each marking a significant shift in how professionals approach data-driven decision-making.</p><p><strong>First Generation: Hard-Coded Programming</strong></p><p>The early days of analytics were dominated by hardcore programming. Data analysts and statisticians relied heavily on scripting languages such as <strong>SAS, R, SPSS, and Python</strong> to write custom code for every analytical task. This approach required deep technical expertise, as users had to manually preprocess data, build models, and interpret results using command-line interfaces and complex scripts. While this method provided flexibility and precision, it also demanded a steep learning curve, limiting data analysis to those with strong programming skills.</p><p><strong>Second Generation: Web-Based Analytical Tools</strong></p><p>As technology advanced, the second generation of analytics saw the rise of <strong>user-friendly analytical tools</strong> that made data science more accessible. This era was marked by the development of <strong>web applications</strong> leveraging underlying programming tools like Python and R but providing a graphical interface for users. Frameworks such as <strong>Streamlit, Dash, Flask, and Guardio</strong> allowed developers to build interactive, web-based analytics applications.</p><p>These applications significantly lowered the barrier to entry, enabling users with minimal coding experience to perform sophisticated analyses. Businesses and researchers could now interact with data through intuitive dashboards, automating many of the manual processes required in the first generation. This shift made analytics more efficient and widely adopted across industries.</p><p><strong>Third Generation: Conversational AI and LLM-Powered Analytics</strong></p><p>Today, we are witnessing the third wave of transformation in analytics, driven by the rise of <strong>Large Language Models (LLMs) and AI-powered conversational interfaces</strong>. In this new paradigm, users no longer need to write extensive scripts or even navigate dashboards manually. Instead, they can simply <strong>ask questions in natural language</strong>, and AI-powered systems generate the required code, execute tasks, and return insights instantly.</p><p>Modern AI-based analytics platforms integrate LLMs with traditional statistical and machine learning techniques, enabling researchers and business users to:</p><ul><li><strong>Automate code generation</strong>: AI tools write and execute Python, R, or SQL scripts based on user queries.</li><li><strong>Perform complex analyses through chat</strong>: Users can interact with AI-driven chat interfaces instead of navigating menus.</li><li><strong>Enhance accessibility</strong>: Non-technical users can now perform sophisticated analytics without needing programming knowledge.</li></ul><p>This shift is redefining the way data analytics is conducted, making it <strong>more interactive, intelligent, and inclusive</strong>. By bridging the gap between human intent and machine execution, <strong>AI-powered analytics democratizes data science</strong>, enabling more people to extract meaningful insights effortlessly.</p><p><strong>The Future of Analytics</strong></p><p>As we move forward, the boundary between human analysts and AI-driven analytics systems will continue to blur. Future advancements will likely bring even more intuitive interfaces, real-time data-driven decision-making, and seamless integration of AI into every aspect of analytics.</p><p>From writing hard-coded scripts to interacting with intelligent assistants, analytics has come a long way. The next phase promises even <strong>greater automation, accessibility, and intelligence</strong>, transforming how we unlock value from data.</p><p>Are we heading towards an era where AI-driven tools will completely replace traditional analytics workflows? Perhaps. But one thing is certain — the way we analyze data will never be the same again.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d36a7cc8a07a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Unlocking the Future: How Multiagent Systems Mirror Human Intelligence Using Advanced AI Concepts]]></title>
            <link>https://venugopal-manneni.medium.com/unlocking-the-future-how-multiagent-systems-mirror-human-intelligence-using-advanced-ai-concepts-8b53abd959b9?source=rss-8f0d88203c4c------2</link>
            <guid isPermaLink="false">https://medium.com/p/8b53abd959b9</guid>
            <dc:creator><![CDATA[Statsvenu]]></dc:creator>
            <pubDate>Sun, 26 Jan 2025 17:02:40 GMT</pubDate>
            <atom:updated>2025-01-26T17:02:40.981Z</atom:updated>
            <content:encoded><![CDATA[<p>In the rapidly evolving landscape of artificial intelligence, one of the most groundbreaking advancements is the development of multiagentic systems. These sophisticated systems emulate human intelligence by perceiving, processing, acting, and learning from feedback. Leveraging the robust principles of Large Language Models (LLMs), Natural Language Processing (NLP), and computer vision, multiagentic systems are advancing towards revolutionary heights. Let’s delve deeper into how they work and their incredible potential for our future.</p><p><strong>Perception: The First Step towards Human-Like Intelligence</strong></p><p>Human beings interact with the world through their senses, gathering and interpreting vast amounts of information. Multiagentic systems begin this process with perception, employing advanced computer vision techniques and NLP to interpret text, images, speech, and video. Through the integration of these technologies, AI can “see” and “understand” its environment with remarkable accuracy, setting the stage for deeper cognitive processes.</p><p><strong>Processing: Transforming Data into Knowledge</strong></p><p>Perception alone is not enough; the perceived information must be processed to extract meaningful insights. This is where LLMs come into play. By understanding the context and nuances of language, LLMs can convert raw data into structured information. For instance, these models can interpret a piece of text, identify key themes, and make inferences, thus mirroring the human ability to understand and analyze language in a comprehensive manner.</p><p><strong>Action: Influencing the World</strong></p><p>The essence of intelligence lies not just in understanding but also in influencing the world. Multiagentic systems use the processed data to make informed decisions and take actions. This could range from answering complex queries to controlling robotic systems in real time. By acting on the processed information, these systems perform tasks with human-like efficiency and precision.</p><p><strong>Learning: The Continuous Improvement Loop</strong></p><p>Learning from feedback is a cornerstone of human intelligence. AI systems are designed to adapt and improve by analyzing the outcomes of their actions. Through iterative learning, where the feedback from actions is continuously integrated into their knowledge base, these systems refine their performance over time. This process ensures that they evolve and improve, mimicking the human capacity for growth and development.</p><p><strong>Integration: The Power of Collaboration</strong></p><p>A significant aspect of multiagentic systems is their ability to decompose larger tasks into smaller, manageable pieces. Think of it as dividing a complex project into specialized tasks, each handled by an expert. In this context, specialized LLMs and other AI components work on discrete sub-tasks, collaborating seamlessly to achieve the overall objective. The collaborative synergy among these agents amplifies the system’s overall intelligence and efficiency.</p><p><strong>Mimicking Human Collaboration</strong></p><p>In the human world, collaboration is key to solving complex problems, and this principle is ingeniously replicated in multiagentic systems. These digital agents don’t work in isolation; they collaborate amongst themselves and with human users. This collaborative framework ensures a holistic approach to problem-solving, blending the strengths of AI and human ingenuity.</p><p><strong>The Future: AI as Collaborative Colleagues</strong></p><p>As we look towards the future, the potential of these systems to function as digital agents or ‘colleagues’ is incredibly promising. Imagine a world where AI systems work alongside humans, providing insights, performing tasks, and offering support, thereby enhancing human capabilities. These agents could transform industries, streamline operations and foster innovation on unprecedented scales.</p><p><strong>Conclusion</strong></p><p>At the heart of these transformative multiagent systems lies the essence of artificial intelligence aiming to mirror human intelligence. By perceiving, processing, acting, and learning, these systems represent a quantum leap in how we approach problem-solving and task management. Their collaborative nature, breaking down complex tasks into smaller components and leveraging specialized AI models, underscores their potential to revolutionize our interaction with technology and reshape our world.</p><p>Embracing this future requires us to stay informed, adapt, and be open to the endless possibilities that AI-driven multiagentic systems bring. As we stand on the brink of this exciting evolution, the collaborative intelligence of humans and AI promises a brighter, more efficient, and innovative future.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8b53abd959b9" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Unraveling “What” and “Why” in Data Science: Phases and Their Importance]]></title>
            <link>https://venugopal-manneni.medium.com/unraveling-what-and-why-in-data-science-phases-and-their-importance-da157e242b50?source=rss-8f0d88203c4c------2</link>
            <guid isPermaLink="false">https://medium.com/p/da157e242b50</guid>
            <category><![CDATA[data-scince]]></category>
            <category><![CDATA[causal-inference]]></category>
            <category><![CDATA[explainable-ai]]></category>
            <dc:creator><![CDATA[Statsvenu]]></dc:creator>
            <pubDate>Mon, 04 Nov 2024 05:51:32 GMT</pubDate>
            <atom:updated>2024-11-04T05:51:32.014Z</atom:updated>
            <content:encoded><![CDATA[<p>In an era dominated by explosive technological advancements, data science has emerged as a cornerstone in deriving valuable insights from massive datasets. As we navigate through its multifaceted nature, it’s crucial to explore the roles of “what” and “why” at different phases of a data science project. This exploration underscores the significance of understanding these components to achieve impactful outcomes.</p><p>Every data science project begins with a clear articulation of the problem statement. This is the most straightforward, Once the problem statement is in place, the next step is data preprocessing — a phase that involves transforming raw data into a clean and usable format. This phase is particularly demanding when it comes to figuring out “what” data is required and how to handle it. This includes:</p><ul><li>Data cleaning and handling missing values.</li><li>Data integration and transformation.</li><li>Feature selection and extraction.</li></ul><p>Given the advent of advanced tools like ChatGPT and GitHub Copilot, the “how” aspect of data preprocessing has become more manageable. These technological aids can automate much of the grunt work, allowing data scientists to focus on deciding “what” to do with the data, leveraging domain knowledge, statistical principles, data literacy skills and analytical experience.</p><p>Modeling represents the core of any data science project. Here, the narrative shifts from “what” to “why.” While setting up and training models involves sophisticated algorithms and advanced AI techniques, it’s imperative to understand the underlying mechanisms — essentially the “why.” Critical aspects include:</p><ul><li>Model choice and justification.</li><li>Understanding model performance and limitations.</li><li>Ensuring model explainability and interpretability.</li></ul><p>In this phase, the emphasis on “why” helps ensure that models are not only effective but also transparent and trustworthy. With sophisticated tools at their disposal, data scientists can delve deeper into causal relationships and underlying patterns, providing robust explanatory frameworks that support decision-making.</p><p><strong>Conclusion</strong></p><p>By appreciating the interplay between “what” and “why” across different phases of a data science project, we can chart a clear path towards meaningful and actionable insights. From the clarity of the problem statement to the precision in data preprocessing and the depth of understanding in modeling, each phase contributes uniquely to the overall success of a project.</p><p>In today’s fast-paced technological landscape, tools like ChatGPT and GitHub Copilot enhance our ability to manage data sciences’ complexities, yet the crux remains — knowing “what” to do with the data and understanding “why” certain decisions and patterns emerge. By maintaining this focus, we elevate our analyses, driving more informed and impactful outcomes.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=da157e242b50" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What is Left to New age data scientist is …]]></title>
            <link>https://venugopal-manneni.medium.com/what-is-left-to-new-age-data-scientist-is-7731424617ee?source=rss-8f0d88203c4c------2</link>
            <guid isPermaLink="false">https://medium.com/p/7731424617ee</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[data-literacy]]></category>
            <category><![CDATA[machine-learning-ai]]></category>
            <dc:creator><![CDATA[Statsvenu]]></dc:creator>
            <pubDate>Sat, 09 Jan 2021 16:05:52 GMT</pubDate>
            <atom:updated>2021-01-09T16:05:52.460Z</atom:updated>
            <content:encoded><![CDATA[<h3>What is Left to New age data scientist is …</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/427/1*TfEkKOfiMByd1ov8tECPYA.png" /></figure><p>Presently we are in the era of automated machine learning (Auto ML), which helps automating most of the machine learning pipeline components such as data understanding, data engineering, feature engineering, model training, hyperparameter tuning, model monitoring, etc. and also there are plenty of such open source Auto ML frame works exists for data scientists to pick and play .so what is left to data scientist is data literacy.</p><p>In my view Machine learning is component of three major components of <strong>Representation===&gt;Learning====&gt; Optimization </strong>, out of which most of the learning and optimization tasks are completely automated by the Auto ML frame works. So we don’t have problem of How to learn and how to optimize due to these Auto ML frame works. The only problem which left to data scientist is how to represent the problem as well as data ,what is the business problem and what are the feature which solve the business problem and most importantly what to optimize ( what is the ideal output of the problem you are looking for and hoe to achieve it ) for this the analyst need to have one key important skill that is called “Data Literacy “ skill, This is going to be the one of the most key differentiator to evaluate the people in future.</p><p>Data literacy skill is the one of the critical thinking, which requires converting the business problem in to data problem as well as creating /finding the relevance data from the big data and also as being machine learning methods are heuristic in nature (There is no hard and fast rules for coding/modelling/performance) and repetitive, we need the data literacy skills to validate the solution and use them</p><p>Therefore for all the upcoming data scientists/Data Analytics please focus more on understanding domain( whatever you are working) and fundamentals of mathematics and statistics along with visualizations and storytelling capabilities as data science is not a code it’s a thought.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7731424617ee" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why we need a mentor in Data science]]></title>
            <link>https://venugopal-manneni.medium.com/why-we-need-a-mentor-in-data-science-b3e5723810ef?source=rss-8f0d88203c4c------2</link>
            <guid isPermaLink="false">https://medium.com/p/b3e5723810ef</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[learning]]></category>
            <category><![CDATA[mentorship]]></category>
            <dc:creator><![CDATA[Statsvenu]]></dc:creator>
            <pubDate>Sun, 27 Dec 2020 13:18:59 GMT</pubDate>
            <atom:updated>2020-12-27T13:18:59.584Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/700/1*ZOhddWO5rDF1zQxbA2LzkA.png" /><figcaption>Why we need a mentor in Data science</figcaption></figure><p>As data science is everywhere due to the evolution of tools, computation power, speed, connectivity, storage systems and algorithms there is no way of getting out of working with it. Due to the pandemic the digitization accelerated furthermore and changed complete Eco system of all the domains. Therefore whatever domain you work for you need provide data driven solutions which serves the customer at his personal level with some kind of smartness around it</p><p>With this reality already people started working towards it and preparing themselves ready but on the process they need a mentor who can direct them in proper direction with the below reasons.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/390/1*QJ31wVOLs2A81Rr6LNYAlg.png" /></figure><p>1) As mentioned above we are in the era of the data explosion , we have plethora of content freely available on data science process, procedures &amp; deployment –etc ,and most of the times people are also getting confused with this content , we really need a mentor who can give connect the content with some kind of context through his real time experience and knowledge</p><p>2) As there is saying that instead of reading some bad books, one good book gives more knowledge in the same lines Good mentor can easily access your capabilities based on that they suggests the related content/areas which you can easily explore</p><p>3) When you are working domain specific projects they always put their real time experience and specific domain experience in your products and makes your more familiar with the domain that will help you to increase your data literacy skills as well as critical thinking.</p><p>4) With his Experience and exposure he will always ensures that you are not deviating from the specific problem and makes you focus on output of the problem you are working for</p><p>5) If your mentor is a practitioner, they can offer a internship project which can give you some real time exposure as well as culture , if not they can find the same in their connections or network</p><p>Therefore here finally, what I am suggesting, whatever way you are learning data science alongside please try to find a mentor with the above qualities by that you will get some kind of direction, prospective as well as discipline in your learning through which you can easily succeed your target.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b3e5723810ef" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Post COVID ‘New Normal’ — Multigenerational point of view]]></title>
            <link>https://venugopal-manneni.medium.com/post-covid-new-normal-multigenerational-point-of-view-f33624331c86?source=rss-8f0d88203c4c------2</link>
            <guid isPermaLink="false">https://medium.com/p/f33624331c86</guid>
            <category><![CDATA[correspondance]]></category>
            <category><![CDATA[association]]></category>
            <dc:creator><![CDATA[Statsvenu]]></dc:creator>
            <pubDate>Wed, 04 Nov 2020 17:01:00 GMT</pubDate>
            <atom:updated>2020-11-04T17:01:00.939Z</atom:updated>
            <content:encoded><![CDATA[<h3>Post COVID ‘New Normal’ — Multigenerational point of view</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/570/0*PUdlAbSxhZI-xxjp.jpg" /></figure><p><strong>Background</strong>: We are in the midst of a unprecedented outbreak of mass proportions. We in India have seen have seen a series of all possible forms of lockdowns and un–lockdowns, the problem still seems to be looming large upon us. A new normal has gotten recalibrated and is here to stay.</p><p>So, what’s on Indians’ minds? Have perceptions, attitudes and beliefs changed with COVID?…how and to what extent?…amongst whom? Let’s figure this out.</p><p><strong>Data source</strong>: Kalos 2020, a worldwide survey across 15 nations, with over 7,000 respondents by Kaleidosense in partnership with JSM (Juxt Smartmandate) Analytics and Toluna. This is India specific data.</p><p><strong>Approach</strong>: Ratings on various attitudes were considered for analysis. The 3 categories of age groups have been defined as, <strong>Gen Z (</strong>aged between 8–23 years), <strong>Millenial</strong> (aged between 24–39 years) and <strong>GenX</strong> (aged between 40–55 years).</p><p>Data was converted to a contingency table of association, as below and used as input for further Analysis.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/240/0*wgW8inztjW9jC4f9.jpg" /></figure><p>We have applied a Technique called Correspondence Analysis (CA) which is a multivariate statistical technique which essentially reduces the dimension of the data without information loss. In other words, CA is used to graphically visualize row points and column points in a low dimensional space. It identifies the main relationships between the rows and columns on a table and plots them on a two-dimensional map. It handles a categorical Independent variable with a categorical Dependent variable</p><p>CA shows relative relationships between &amp; within two groups of variables. Relative distances &amp; angles between and within variables &amp; from origin, are key to drawing insights</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/300/0*BlgsgYyPUTezcbs1.jpg" /></figure><p><strong>Insight</strong></p><p>With the above correspondence analysis map, we arrived at more apparent conclusions and it allowed us to quickly identify the associations</p><p>GEX X (Aged 40–55) is now anxious about queuing with strangers and scared about sneezing in public; is beginning to respect and value the role of those who stay at home and do house work; does not mind paying more eco-friendly products and feels that Biologicals(virus) are the new WMDs (weapons of mass destruction) and the world does not need weapons anymore!</p><p>Whereas Millennial (age 24–39) feel that their anxiety levels are up, dreading the thought of wearing masks all the time in future, and suddenly see that wearables are not so useful as thought out to be. This generation is in fact is very troubled with ‘wfh’(working from home) and feels that family time is overrated J ; are of the opinion that Cinema will lose relevance with so much content being available online.</p><p>Whereas Young kids GENZ (Aged 8–23) has a ‘carpe diem’ philosophy and agrees that one has to live for now (because you never know what tomorrow has in store for you); and of course, ‘friends are the key to life’;</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f33624331c86" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Post COVID ‘New Normal’ — Attitudinal Segmentation]]></title>
            <link>https://venugopal-manneni.medium.com/post-covid-new-normal-attitudinal-segmentation-33e7c43aaae9?source=rss-8f0d88203c4c------2</link>
            <guid isPermaLink="false">https://medium.com/p/33e7c43aaae9</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[analytics]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[cluster-analysis]]></category>
            <category><![CDATA[segmentation]]></category>
            <dc:creator><![CDATA[Statsvenu]]></dc:creator>
            <pubDate>Wed, 04 Nov 2020 16:57:16 GMT</pubDate>
            <atom:updated>2020-11-04T16:57:16.485Z</atom:updated>
            <content:encoded><![CDATA[<h3>Post COVID ‘New Normal’ — Attitudinal Segmentation</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/570/0*JviAG10l8r-DoRxz.jpg" /></figure><p><strong>Background</strong>: We are in the midst of a unprecedented outbreak of mass proportions. We in India have seen have seen a series of all possible forms of lockdowns and un–lockdowns, the problem still seems to be looming large upon us. A new normal has gotten recalibrated and is here to stay.</p><p>So, what’s on Indians’ minds? Have perceptions, attitudes and beliefs changed with COVID?…how and to what extent?…what are the key segments that emerge based on the attitudes and beliefs</p><p><strong>Data source</strong>: Kalos 2020, a worldwide survey across 15 nations, with over 7,000 respondents by Kaleidosense in partnership with JSM Analytics and Toluna. This is India specific data.</p><p><strong>Approach</strong>: Ratings on various attitudes were considered for analysis, these attitudes related to Quality time, learning new things, Retail Experience, WFH, outdoor activities..Etc. . Check below for the list of attitudes considered for the analysis.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*oYREFYnXBBa-GyqB.jpg" /></figure><p>These inter-correlated attributes were condensed to a number of composites using data reduction techniques (factor analysis) and these composites were then used to create clusters/segments using K means algorithm. Once find the appropriate clusters /segments and then these segments further examined and profiled</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*oBZ-0V2ovgDmfpoF.jpg" /></figure><p><strong>Impact</strong>: From the segments and their profiles, the below are the different segments emerged from the attitudes/beliefs about Covid-19 and the derived segments are</p><ul><li>Pessimist (27%)</li><li>The Wise/Silver Gen (27%)</li><li>Masked Warriors (15%)</li><li>Urban Survivor (30%)</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=33e7c43aaae9" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What is causing Increasing Anxiety Levels in this Pandemic time?]]></title>
            <link>https://venugopal-manneni.medium.com/what-is-causing-increasing-anxiety-levels-in-this-pandemic-time-b4b57c075bf8?source=rss-8f0d88203c4c------2</link>
            <guid isPermaLink="false">https://medium.com/p/b4b57c075bf8</guid>
            <category><![CDATA[logistic-regression]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[covid19]]></category>
            <category><![CDATA[analytics]]></category>
            <category><![CDATA[supervised-learning]]></category>
            <dc:creator><![CDATA[Statsvenu]]></dc:creator>
            <pubDate>Wed, 04 Nov 2020 16:53:21 GMT</pubDate>
            <atom:updated>2020-11-04T16:53:21.459Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/570/0*9-6cpCxePh1VWRpR.jpg" /></figure><p><strong>Objective: </strong>As the world is witnessing an unprecedented outbreak of mass proportions, and all of us are doing our bit by ensuring ‘social distancing’, essentially locking ourselves inside our own homes, totally avoiding social interactions. We wanted to understand what are the different attitudes/actions which are causing increased anxiety levels in this pandemic.</p><p><strong>Data source</strong>: we have considered India data from the Kalos 2020 a survey done at an international level across 15 nations around 7,000 respondents by Toluna &amp; Kaleidosense.</p><p><strong>Approach</strong>: As in the Kalos 2020 respondents were asked to rate 4 point scale on various attitudes to life change due to covid , such as Quality time, Learning new things, Retail Experience, WFH , outdoor activities..etc. and also captured the individuals anxiety levels . In order to find what is driving increased anxiety levels we categorized the increasing anxiety levels into positive and negative and find the drivers what is causing positive in increased anxiety levels. This analysis we have done at overall level as well as among males and females to check is there any differences emerge.</p><p>As being Increased Anxiety levels is being dichotomous (positive/negative) variable, we have applied a popular well known classification technique call logistic regression and regressed this increased anxiety levels on various Quality, Learning new things , Retail Experience, WFH, Outdoor activities related attributes and selected the statistically significant attributes which are driving the increased anxiety levels with the accuracy of 90%.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/724/1*hYKRM64o8ATaXS5v0b3krw.png" /></figure><p>Then we have normalized the each attitude importance from the above table and then calculated relative importance through which we can order these attributes as per their order importance on causing increased levels of anxiety. And the below is the same for at an overall level as well as among males and females.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*vh_dvAeJ_-c6L0xt.jpg" /></figure><p><strong>Impact</strong>: From the above. At an overall level we can see that somehow Quality family time is over rated than actual causing more anxiety followed by not able to enjoy the earlier retail experience and not able to enjoy the routine life ,due to not able spend more time outdoor and scary of Coughing/Sneezing in public and also noticed there is clear difference for increasing anxiety among males and females</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b4b57c075bf8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why do we have so many modelling algorithms]]></title>
            <link>https://venugopal-manneni.medium.com/why-do-we-have-so-many-modelling-algorithms-9a1c6a125485?source=rss-8f0d88203c4c------2</link>
            <guid isPermaLink="false">https://medium.com/p/9a1c6a125485</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[analytics]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[algorithms-and-data]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Statsvenu]]></dc:creator>
            <pubDate>Wed, 04 Nov 2020 16:48:37 GMT</pubDate>
            <atom:updated>2020-11-04T16:48:37.834Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/570/0*mbXQWtFMDiREhbsk.jpg" /></figure><p>As a practitioner, in my lens any of the analytical problem will be solved by using the either of these approaches</p><ul><li>Prediction — Predicting the future</li><li>Grouping</li><li>Association</li><li>Exploring the data in either higher dimensional Space (kernels) or lower dimensional Space</li></ul><p>And we do have different algorithms because of the variable type of what we are predicting, in case of continuous prediction we have regression algorithms, in case of discrete or nominal variables we have classification algorithms and if you want to predict the future by using past variables we have time series algorithms.</p><p>Whereas when it comes to grouping, in case you are interested to group the variables we have data reduction techniques and if you want to group the observations we have cluster algorithms.</p><p>Similarly for the Associations we have Market Basket Analysis or recommendation systems which are based on either content as content based recommendation systems or if you want to provide recommendations based on finding similar people we have collaborative filtering.</p><p>All these basically learning based approaches and we do have different ways of finding these relations using different approaches the below are different approaches and models that exists to solve the analytical problem which might fall under prediction, grouping and association categories.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*rMjBE5ggx9IzrjPp.jpg" /></figure><p>As there is “<strong>No Free Lunch” theorem</strong> states that there is no one model that works best for every problem. The assumptions of a great model for one problem may not hold for another problem, so it is common in machine learning to try multiple models and find one that works best for a particular problem. That’s the reason we practitioner always build the pipeline of algorithms and select the best model on the basis of cross validation in both external(domain) and internal(data) prospective</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9a1c6a125485" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[WHY AI is SOCIAL GOOD Technology]]></title>
            <link>https://venugopal-manneni.medium.com/why-ai-is-social-good-technology-b1171590889a?source=rss-8f0d88203c4c------2</link>
            <guid isPermaLink="false">https://medium.com/p/b1171590889a</guid>
            <category><![CDATA[machine-learning-python]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[machine-learning-ai]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Statsvenu]]></dc:creator>
            <pubDate>Wed, 04 Nov 2020 16:45:14 GMT</pubDate>
            <atom:updated>2020-11-04T16:45:14.202Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/570/0*lXm1OWczCVNGm062.png" /></figure><p>I believe that ‘technology/ AI’ does not have intrinsic value; its value lies in its application. I always say that I may create the best of algorithms with a great accuracy but it is only as useful as the problem it is solving. When applied critically to certain social problems, technology can help to facilitate significant, positive social change.</p><p>“AI for social good”. We have been seeing large organisations like<a href="https://ai.google/social-good/"> Google</a>,<a href="https://www.microsoft.com/en-us/ai/ai-for-good"> Microsoft</a> and Facebook talk about this.</p><p>“Applying AI to some of the world’s biggest challenges. Through research, engineering, and initiatives to build the AI ecosystem, we’re working to use AI to address societal challenges.”<strong>-Google</strong></p><p>“AI for Good. Providing technology, resources, and expertise to empower those working to solve humanitarian issues and create a more sustainable and accessible world.” <strong>-Microsoft</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*bMeL8SO8EkHwQn1V.png" /></figure><p>“Facebook has launched a host of initiatives to deliver social and inclusive growth by harnessing the power of emerging technologies led by artificial intelligence (AI)”.<strong>-Facebook</strong></p><p>I attended a couple of virtual conferences, one was <a href="https://raise2020.negd.in/community">RAISE 2020</a> (Responsible AI for Social Empowerment) organized of Government of Indian in collaboration of Ministry of Electronics and information technology, NITI Aayog and Digital India. The other was <a href="https://nasscom.in/aisummit/">NASSCOM Xperience AI summit 2020</a></p><p>The priority areas identified were Healthcare, Agriculture and Education. It was wonderful to see necessary push and support coming from the government, the strong participation from the private sector and the enthusiasm and passion shown by the start-ups. It takes this 360 degree involvement for products and applications to reach and benefit the common man.</p><p>The ideas ranged from personalisation in education and precision medicine to predictive applications that help take proactive measures that help mitigate negative outcomes.</p><p>What aids this in India, is our plummeting data prices. According to a 2019 BBC report, <a href="https://www.bbc.com/news/technology-47416250">we have the cheapest mobile broadband prices in the world</a>. In addition, given our population, India has a tremendous potential for technological advancements.</p><p>By 2025, just in another 5 years, AI technology will become so consumer centric, will have percolated across social strata, will be accessible common man, will provide inputs and guide in the most personalised level in the 3 most crucial areas of health, education and agriculture. This will be true realisation and translation of the phrase “AI for social good”</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b1171590889a" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>