<?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 Jesse Mondragon on Medium]]></title>
        <description><![CDATA[Stories by Jesse Mondragon on Medium]]></description>
        <link>https://medium.com/@mondregon?source=rss-bb0f6ff971c6------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*W17bh7AN0bPmM-sT.</url>
            <title>Stories by Jesse Mondragon on Medium</title>
            <link>https://medium.com/@mondregon?source=rss-bb0f6ff971c6------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 00:35:57 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@mondregon/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[Adamant.]]></title>
            <link>https://medium.com/data-science/adamant-3d87e8d98b83?source=rss-bb0f6ff971c6------2</link>
            <guid isPermaLink="false">https://medium.com/p/3d87e8d98b83</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[predictive-analytics]]></category>
            <category><![CDATA[predictive-modeling]]></category>
            <category><![CDATA[food]]></category>
            <dc:creator><![CDATA[Jesse Mondragon]]></dc:creator>
            <pubDate>Fri, 28 Apr 2017 16:20:21 GMT</pubDate>
            <atom:updated>2017-04-28T16:53:49.678Z</atom:updated>
            <content:encoded><![CDATA[<p>Hi again! My internet was inexplicably acting up yesterday and I did not get the chance to write a bit about a revelation I had! The last couple of days I have been putting together data for the project with <strong>InCahoots </strong>and I have found a new method to approach it using predictive analytics. It not necessarily new or new to me, but sometimes when you start working on a project you can remain so adamant to stick to one way that you forget that there are many angles to tackle on a project.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RGpltFdhsqkZWkHJFTSCMg.png" /></figure><p>Above is a snippet CSV file that I created that includes each venues features. For now, we are experimenting with each feature and how well they predict customer preferences. Back to the revelation, the past couple of days have just been <strong>web scrapping </strong>days where I am extracting information (customer reviews, ratings, and comments) from each venue’s website and each of their different social media pages. A few features are your typical ones — venue type, ambience, menu style, location, etc — but then I decided to have some fun.</p><h3><strong>Pokémon</strong></h3><p>My first “for fun” test the correlation between <em>nearby Pok</em>é<em>mon stops </em>and the likelihood a customer will go to that venue. It may or may not have any significant correlation or business value, however, if there is even the slightest possibility that this feature attracts customers, I am going to test for it. I know I am going again the golden rule in statistics “<em>correlation does not imply causation,” </em>however, it is a fun project and if it brings back anything valuable, I will share it to you all ( I think I just broke another golden rule… the Texan rule of saying ya’ll instead of you all… whatevs).</p><h3><strong>Similarity Test</strong></h3><p>OK, back on track. So with my given set, I am going to perform a Cosine Similarity/Similarity Analysis test (similar to what Amazon and Netflix does to recommend similar products to us). However, before I do get started, I have to expand my dataset and include those venues that have not been visited yet, in order to get a ranking system created and see which businesses rank the best in similarity with the venues that are already signed.</p><h3><strong>Machine Learning</strong></h3><p>My initial approach was to use the discrete variables to classify the different venues — signed and unsigned — and then examine whether or not this would be a good predictor/indicator if whether a venue would do business with <strong>InCahoots.</strong></p><p>The leadership team at <strong>InCahoots </strong>has a theory that a strong social media presence equate to ideal venues to work with; so give this information I was also going to run a Logistic Regression (LogReg) test the validity of this hypothesis. I have split my datasets by those that have signed, and potential venues (my train and test datasets). From there, I will be using the features that I have related to social media, on the training set, and calculate whether a bar will sign or not.</p><h3>Web Scraping</h3><p>The web scraping will allow me to extract comments and build a text analyzer to really get to understand the customer experience and distinguish each venues strengths. I used the library <strong>BeautifulSoup</strong>, a go to library, to extract the content from the different websites. Below is the first step, before I get to the nitty-gritty part. Right away I can see that the majority of the comments have to do with the breakfast options. The next step is to further analyze the rest of the comments and really make smart and appropriate decisions with this information.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/503/1*_tc7B9bjNajZNkDdxS_zcg.png" /></figure><h3><strong>Conclusion</strong></h3><p>I know I did not go too much into detail on my methodology, mostly because I will write individual blogs for each different approach and go further into detail. For now, this is the overall plan. Working with a small start up is hard work, but extremely beneficial. It has challenged me to use all of my knowledge to tackle on each task from a different point of view. The more information I have, the more confident the leadership team feels about their business. I will be back in the next couple of days with an update on the project.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3d87e8d98b83" width="1" height="1" alt=""><hr><p><a href="https://medium.com/data-science/adamant-3d87e8d98b83">Adamant.</a> was originally published in <a href="https://medium.com/data-science">TDS Archive</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Starting Up pt. 2]]></title>
            <link>https://medium.com/the-ascent/starting-up-pt-2-d3b9ccd1a0ea?source=rss-bb0f6ff971c6------2</link>
            <guid isPermaLink="false">https://medium.com/p/d3b9ccd1a0ea</guid>
            <category><![CDATA[python]]></category>
            <category><![CDATA[exploratory-data-analysis]]></category>
            <category><![CDATA[data]]></category>
            <category><![CDATA[business]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Jesse Mondragon]]></dc:creator>
            <pubDate>Fri, 28 Apr 2017 13:02:43 GMT</pubDate>
            <atom:updated>2017-04-28T16:06:23.562Z</atom:updated>
            <content:encoded><![CDATA[<p>So far, some of the projects that I have been working on using Data Sciences have all been business related. Since I am working with a Start up (I will refer to it as <strong>InCahoot, </strong>I’m not sure why I chose that name) that has not yet launched, I have to be creative with the research and reports that I present to the CEO. Due to privacy constraints, I am limited in what I can share, but I will try to be as detailed as possible.</p><p>My<strong> first step </strong>was to create, together with the CEO, an objective listing all of the areas where the company can grow using analytics.</p><ol><li>They were using what was described as a <strong>shot gun approach </strong>to sign new clients — their clients are restaurants, bars, breweries, and pubs. The company has a product that needs these venues. Although they have been able to sign about 15 businesses in the Dallas area, the CEO believes that their approach of randomly choosing a place and emailing them is time consuming and not well thought out. So, my first task was to find a solution that could save them both time and money.</li></ol><p>I created a list of things they need: customer data, venue data, and demographic data (according to zip code). My end goal was to create a profile of the average consumer that would fit our ideal profile. I had to get census data in order to achieve this goal and from there use inferential statistics to get a 5 number summary (minimum, first quartile, median, third quartile, and maximum). My 5 number summer included the following categories: age, salary, education, renter or home owner, ethnicity, sex, marital status, and a couple of more categories.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/679/1*yAz5yTBannXxDj_Z_Q1Pkw.png" /></figure><p>I chose zip codes as my parameters because with my experience in dealing with city data I have seen that it’s a pretty good indicator of the population that reflects a businesses average consumer.</p><p>My next step was to start doing some exploratory data analysis, using python, and start putting together some descriptive statistics for a quick review of the data collected.</p><p><strong>Below are a couple examples:</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/886/1*sykBRjnov026V4wtr_0-yA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/901/1*HENiaNMP7KhNHqE4Hh2LGA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/923/1*t2yW5sl3PrtGwvj2Opt1zg.png" /></figure><p>The next step is to start seeing the frequencies and getting percentages to get an idea of the averages and figure out who lives in which zip code and how accurately does that reflect the businesses average consumer. My next blog will have more details on the second objective and how I plan to tackle on the next problem.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d3b9ccd1a0ea" width="1" height="1" alt=""><hr><p><a href="https://medium.com/the-ascent/starting-up-pt-2-d3b9ccd1a0ea">Starting Up pt. 2</a> was originally published in <a href="https://medium.com/the-ascent">Ascent Publication</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Starting Up]]></title>
            <link>https://medium.com/@mondregon/starting-up-77af4d414fa3?source=rss-bb0f6ff971c6------2</link>
            <guid isPermaLink="false">https://medium.com/p/77af4d414fa3</guid>
            <category><![CDATA[internships]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[big-data]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[newbie]]></category>
            <dc:creator><![CDATA[Jesse Mondragon]]></dc:creator>
            <pubDate>Fri, 28 Apr 2017 12:11:22 GMT</pubDate>
            <atom:updated>2017-04-28T14:20:07.225Z</atom:updated>
            <content:encoded><![CDATA[<p>A lot has gone down since my first blog: I got an internship as a Data Scientist and it has picked up pretty fast! It being my first opportunity to exercise what I have learned has been very beneficial when it comes to acquiring that much needed experience. So first thing is first, before I get to any technicalities, I am going to briefly talk about how I came across this opportunity for those of you who are also transitioning to Data Sciences. If you wish to skip this part, which gets a little lengthly, and are more interested in what my job is asking me to do, I would skip to the second part of my blog <strong>Starting Up pt. 2.</strong></p><p>I met my current my boss in a serendipitously manner. My roommate received a text from a friend who was getting ready to launch his own start up. He wanted my roommate go like his Facebook page, which my roommate did and as he was doing, he was telling me what was going on. I jokingly told him, “hey, you should tell your friend about me *laughs*.” My roommate did, which to my surprise, his friend was excited to meet me and set up a lunch with us right away. I was surprised that it had worked, but very excited. Right from the start, my now boss communicated with us that he is not making any money right now since the business has not launched yet. Of course, that did not matter to me because all I was looking for was to speak to him and see if I could work with him on a project so that I can get a project under my belt (my current Data Sciences program requires us to do a capstone to complete the course). Honestly, I was undershooting. The lunch went great, and everything he touched on as far as his companies needs and how an analytics team could help, was right on point with what I am looking for as far as my role for a company. He extended me an offer to jump on board as an intern and I did not hesitate. This is what I want to do.</p><p><strong>(ADVISORY: I go on a complete tangent below)</strong></p><p>I had spent the last 12 years wandering around looking for the perfect situation. The perfect job. The perfect paycheck. The only problem? A younger me was always afraid to take risks and work hard. I had played it safe by dropping out of college when I was 19 and working as a construction worker, landscaper, and installing carpets in student apartments; which was all hard work, but not smart work. When I returned to UT, I still played it safe, and transferred into a degree I was not passionate about, rather, one that I felt complacent in doing. This has made the journey worth it, thus far.</p><p><strong>ANYWAYS, those of you out there who are also struggling, my advice is to be very proactive. Go to meet ups, networking events, talk to people you know (they may know someone who knows someone). Don’t be afraid to work for free aka internships (this mostly goes for students; I know that this may not be ideal for everyone).</strong></p><p>My next step is to apply to companies that will have me do similar projects like the ones I am currently doing for my internship. Honestly, I will continue to do projects and Data Science work even if I get a job as a Data Analyst. My ultimate goal is to be a great Data Scientist because I love what it entails. I strive to find the unknown and showcase it to people.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=77af4d414fa3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Taking L’s]]></title>
            <link>https://medium.com/@mondregon/taking-ls-64337b9c7de?source=rss-bb0f6ff971c6------2</link>
            <guid isPermaLink="false">https://medium.com/p/64337b9c7de</guid>
            <category><![CDATA[coding]]></category>
            <category><![CDATA[education]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[analytics]]></category>
            <dc:creator><![CDATA[Jesse Mondragon]]></dc:creator>
            <pubDate>Thu, 13 Apr 2017 14:32:55 GMT</pubDate>
            <atom:updated>2017-04-13T14:32:55.266Z</atom:updated>
            <content:encoded><![CDATA[<p>By definition, I am a newb in the field of data science. I do not have a solid STEM background, mostly because I feel like I never that the opportunity to. By that I mean that my K-12 education did not have the resources that my fellow classmates and I could take in and nurture that love for STEM that I saw many times in college. Nonetheless, that’s a story for another time; today I am going to share my story and hopefully it reaches someone out there that is thinking about a career into data, or a young person that is looking into the field but is unsure. Please excuse any run- on sentences or any other missing punctuation, I am writing what ever comes to mind and sometimes it sounds great in my head but not so much on paper (or blog, in this case).</p><p>As I mentioned, I did not have a great start in the tech field. I started my freshman year in college at The University of Texas at Austin as a Computer Science Major (CS). Coming out of a low- income and resource community that was East Austin, back in the day, I always felt behind. As a young person fresh out of high school, I think, like many people, I got intimated and never gave myself a chance in the field of CS. Very quickly I determined that I just did not want to do programming and I quit it entirely—well, flash forward 10 years to present day and I find myself going full- circle. Today, the field of analytics and data sciences has got my attention. I stumbled upon this passion serendipitously in a class that I loathed: Statisitcs. It was mandated in my degree course to take a math course, and I hated math. I left the CS field in 2009, and this was 2015, but I still hated math. I still had that fear of it. That fear of failure. It was one of the reasons why I transfered out of CS. However, this time it was different. I don’t know if it was because I was older and had picked up better study habits, or if it was because I just had a different mentality, whatever it was, I found myself really loving statistics. What was ever better was that this course applied those statistics to real world problems, it wasn’t just another theory class. After all my loses and insecurities, I saw math with a new perspective.</p><p>I know I kinda took you through a wild loop and added details that may or may not be interesting, but I just want those of you who are in a similar situation that I was in to know that you are not alone. If you are anything like me and you are determined to find you passion but you feel unsure of yourself sometimes, just remember the long- term benefits. I learned that all of my fear did nothing to stop me from wanting to work in the tech field, however, each time I did cave in to that fear, I had to start all over. That’s what I mean about thinking long term. Give yourself a change to struggle in this field, it is not an easy one but there are many other people out there too that are in the same position.</p><p>Enough rambling, if you would like me to keep elaborating on my journey to where I came from, I can also submit anecdotes frequently just so you can see how many mistakes I have made but how I have grown from them; whether it pertains to the field of data or just in general how I went from a kid who everyone thought would be a high school drop out and a druggy, to someone that just finished their Master’s at UT.</p><p>For future blogs, I will start posting more on my work in the industry and how I want to bring in my personal style to it. I know I am still a novice, so take it easy :D but I am also determined to learn something new everyday and grow in the field.</p><p>Thanks,</p><p>JM</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=64337b9c7de" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>