<?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 Letícia Ange Pozza on Medium]]></title>
        <description><![CDATA[Stories by Letícia Ange Pozza on Medium]]></description>
        <link>https://medium.com/@leticiapozza?source=rss-c76f7d1baba8------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/2*xR6FMklLksA1CaQbe0kH4w.jpeg</url>
            <title>Stories by Letícia Ange Pozza on Medium</title>
            <link>https://medium.com/@leticiapozza?source=rss-c76f7d1baba8------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 21:44:54 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@leticiapozza/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[Resilience is key in data science: supporting SaMi for Brazilian’s health system]]></title>
            <link>https://medium.com/weareodd/resilience-is-key-in-data-science-supporting-sami-for-brazilians-health-system-3cd009549fc8?source=rss-c76f7d1baba8------2</link>
            <guid isPermaLink="false">https://medium.com/p/3cd009549fc8</guid>
            <category><![CDATA[data-scientist]]></category>
            <category><![CDATA[data-product-creation]]></category>
            <category><![CDATA[public-policy]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[big-data-in-healthcare]]></category>
            <dc:creator><![CDATA[Letícia Ange Pozza]]></dc:creator>
            <pubDate>Wed, 29 Jul 2020 19:00:28 GMT</pubDate>
            <atom:updated>2020-07-29T22:35:34.153Z</atom:updated>
            <content:encoded><![CDATA[<blockquote>This is a summary of our learning from one of many projects we support at the Bill &amp; Melinda Gates Foundation for the GCE in data science and health, and a series talking about these project’s products and results. We will be posting about the others as their results are published. We hope you enjoy!</blockquote><h3>About the Grand Challenges Explorations and our work supporting projects from the Gates Foundation</h3><p>For the past year <a href="https://odd.studio/">Odd.Studio</a> has worked with more than 14 projects in data science with brilliant researches in public health, computer science, epidemiology and many other related areas. Our company is part of a bigger strategy group that supports these diverse research teams that have received USD 100,000 grant from the Bill &amp; Melinda Gates Foundation (BMGF) and from the Ministry of Health in Brazil, to look at ways to improve maternal and child health in Brazil, leveraging and exploring available data. Brazil has huge administrative databases in health, one of the results from having a unified health system in the country.</p><p>The Grand Challenges Explorations in Data Science is a program from the BMGF that started in 2018 and is working to build a community in data science for health. We have 14 projects in its final stages in Brazil, selected from the first round, and 20 others starting in India and Africa. Soon, we will be an ecosystem of ~250 researchers connected worldwide, with incredible knowledge in data sources from all these locations, data science analysis techniques and machine learning models, treatment and analysis of huge volumes of data, and many other related topics.</p><p>All of it was built from a local perspective, with stakeholders involved from the beginning, helping define the topics that were eligible for propositions, connecting to local data partners and sources. Thus, creating the potential to impact the lives of millions of women, mothers and children in these locations, as well as to support decision-makers and guide local stakeholders to learn and use this deliverables and results, shaping public policy.</p><h3>SaMI’s Journey: Interacting with Stakeholders to build a Data-Driven Platform to Support Decision-Making</h3><p>One of the selected projects is from Federal University of São Paulo in collaboration with Unicamp, where a team of researchers has built a platform (<a href="http://prainha.cti.gov.br:1333/">SaMI</a>) to support decision-making, that among other visualisation tools and models, could predict the probability of survival of a newborn on its first 28 days of life, based on a list of atributes from the baby, the mother, and its location. To build the 94% accuracy model, they used more than 40 million data samples from 2006 to 2016. After running the prediction, these variables are ordered to indicate which of them had the biggest impact on its outcome. You can run the model <a href="http://prainha.cti.gov.br:1333/prediction">here</a> (only in Portuguese), but they have a publication about it <a href="https://www.sciencedirect.com/science/article/pii/S2352914820302112?via%3Dihub">here</a> in English.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*H7N_qmP3xOgGUWSgVr69ww.png" /><figcaption>Here is an example from one of the graphs explaining the model, indicating what had the biggest impact on the risk of neonatal death of the inputs I have given the model. The red indicates that the conditions selected in which the newborn would be born were not ideal, and that it has 94% probability of not surviving in its 28 first days, being the Apgar and the Robson Classification the main reasons for this outcome.</figcaption></figure><p>But even though this is an amazing tool, when presented to our local stakeholders — technicians that were going to consume this data to plan strategies to reduce neonatal mortality in Brazil — , the information was aggregated in a way that made it difficult for these group of people to carry it forward or to even analyse it. <strong>They were not ready to use that type of information, at least not at that moment.</strong></p><h4><strong>It is actually pretty common for data science teams that have been working months on something not to realize that they are a lot of steps ahead of the user’s ability to consume it. There are a number of reasons for it to happen: not enough time to touch base, the request was not specific, the data doesn’t answer that specific request…</strong></h4><p>And this is where the beauty of the GCE program comes in: when discussing the possibilities we understood that the model they needed was much simpler, in a way, but just as important. When digging into their necessities on predicting neonatal mortality in Brazil one of the experts pointed out that predicting itself was not the problem, but looking at it in real-time, was. Apparently, there is a 4-month-gap from neonatal deaths occurring in many areas in Brazil and the data they look at. So what they thought was a real-time problem, we understood as a predictive opportunity.</p><p>Why not use data science to predict where and when neonatal mortality rates would rise, trying to close the gap between the information they had access to and what would happen in the near future? With that information, they could plan 1, 2, 3 or 4 months ahead understanding where they should focus on. Something only possible when interacting with stakeholders and consumers of this type of information.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*liDbt2N97r2bF9L8lL0-qQ.png" /><figcaption>Other research groups had already identified that there is a seasonal component to neonatal mortality in Brazil (in some regions more than others). So when the opportunity came, we were pretty confident that by using temporal series analysis they would be able to predict neonatal mortality with an acceptable accuracy, specially for the intended purpose. The continued lines are the predicted rates for months ahead (red for 1 month, purple for 2, blue for 3 and yellow for 4) in a specific region in Brazil — Aracaju. The green, dotted line represents the real data. We’ll write more about this seasonal component when the results are officially published.</figcaption></figure><p>And yes, we know that we are privileged to have access to the final user because of the program. Some researchers don’t even know where to start when we talk about accessing and considering their final user in their process, and others are more interested into writing their papers. This is one of the reasons we are supporting this community: to help translate their work into deliverables that are easier to be understood by local stakeholders. So making sure we create this conversations and help translate to both parties is part of our role with the Foundation.</p><h4><strong>But how many ARE close to the final user and don’t even bother to include them in the development process? This leads to lots of projects being completely put aside, due to not commiting to small changes on data science project/product that could have real impact on a country’s population. It is not only about asking the right questions, it is about actually listening to the answers.</strong></h4><p>And for that, I completely applaud SaMI’s team and their spirit: they had the opportunity to make something useful and they took it. Even if it meant reevaluating the efforts and resources from their original plan, its importance to the greater public was put first. What they had done so far laid ground for them to see a way of using data for good, and they had the knowledge to act fast on it, changing mid project without any doubt.</p><h3>Being resilient when interacting and listening to your end user: a key competence for data scientists</h3><p>I’ve seen many data scientists wishing to be seen and to be heard, and to develop THE project that will change their careers, but as long as it means applying the craziest techniques and doing it on their terms.</p><p>I see a smaller group that has understood that data science is a support area, and that if you are not a good listener and a team player your project will probably remain in a drawer. SaMi’s team was opened from the beginning and supporting their final products has been a huge pleasure and learning opportunity for us. They’ve reminded us that:</p><ul><li>Data science is about understanding your role and playing your strengths, but being humble to accept (and ask for) help in areas that you don’t have a lot of experience;</li><li>Science can be applied to policy making when both parties are ready to contribute and listen, but sometimes, being present requires letting go of personal beliefs and objectives to support a common, greater, goal;</li><li>The importance of a “translator” that helps with a systemic view can make things quicker and leverage opportunities that specialists tend not to see due to being deeply involved into their project’s details;</li><li>Visualising and showing your results in a compelling and simple way facilitates conversation between diverse areas or knowledge.</li></ul><p>Our special thanks to Tiago Carvalho and their whole team for putting up with all our requests and nagging over these 18 months. We hope to see more of this project’s amazing results in the future, and you can always count on us to make it happen.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3cd009549fc8" width="1" height="1" alt=""><hr><p><a href="https://medium.com/weareodd/resilience-is-key-in-data-science-supporting-sami-for-brazilians-health-system-3cd009549fc8">Resilience is key in data science: supporting SaMi for Brazilian’s health system</a> was originally published in <a href="https://medium.com/weareodd">oddstudio</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Creating Odd.studio: Applying the Mesa Method to build our own brand]]></title>
            <link>https://medium.com/weareodd/creating-odd-studio-applying-the-mesa-method-to-build-our-own-brand-c7695351b169?source=rss-c76f7d1baba8------2</link>
            <guid isPermaLink="false">https://medium.com/p/c7695351b169</guid>
            <category><![CDATA[data-strategy]]></category>
            <category><![CDATA[design-thinking]]></category>
            <category><![CDATA[design-process]]></category>
            <dc:creator><![CDATA[Letícia Ange Pozza]]></dc:creator>
            <pubDate>Mon, 02 Mar 2020 19:31:01 GMT</pubDate>
            <atom:updated>2020-03-02T19:31:01.327Z</atom:updated>
            <content:encoded><![CDATA[<p>In May 2019 we started our company Odd.Studio. It was a wish and an can’t-miss type of opportunity, all in one. From May to December we worked and traveled a lot, so there was little to no time to talk about branding. We were having an amazing opportunity to test ourselves as entrepreneurs before actually translating what we thought we were and want to represent. Not every company has that privilege, so we wanted to take our time, building portfolio and getting to work as a team for the first time.</p><p>In November, we decided we were ready to start discussing our brand, but we decided to do it ourselves, and not just participate in the process. After all, our vision was to build a studio with graphic design, experimental and practical design-like interventions with our clients, and this was the perfect opportunity to test our abilities. <strong>Hiring some other company to do it did not feel right, so we understood it was the perfect opportunity to test the Mesa Method.</strong></p><h3>But what is a Mesa and why use it as tool?</h3><p><a href="http://mesa.do/">Mesa</a> is a Brazilian company that has reinvented problem solving and how consulting works. They apply a 5-days sprint method to come up with a real deliverable at the end of day 5. Their method comes from the belief that a <em>Mesa (“table” in Portuguese)</em> is the perfect combination of work and pleasure, and all you need to work is talented people around a table to solve real and complicated problems. I have already been asked to contribute in 2 <em>“Mesas”</em> and not only have enjoyed it and learned a lot from it, but wanted to test my abilities as a leader, applying their method to solve our problems <em>(if you want to learn more, please go to mesa.school, they have an online course that teaches their method, and although participating in one teaches you way more, I do recommend it).</em></p><h4>The method was perfect for us: we didn’t have a lot of time, but wanted to do something awesome that captured who we were to start 2020 with a logo, portfolio and a digital representation of our work. So we set out a budget, made sure we would take a week off just to focus on it, and started prepping for the process.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vkY70Y9MR9TqaJwuYi1w1w.jpeg" /><figcaption>Setting up a Mesa is an important ritual, from choosing the teams and where they seat!</figcaption></figure><h3>Writing your Mission and choosing your Team</h3><p><strong>are the most important parts of getting a Mesa right</strong></p><p>You never know exactly what is going to come out at the end of a Mesa, and that is the excitement of it, but you have an idea of the professionals needed because of the Mission. Building your team and writing a good Mission are key to having a good sprint, and we definitely did that: it is a great opportunity to really test someone’s abilities as makers and by the end of Day 2 you are so aligned in vision, ideas and goals with everyone around you that even though you are working simultaneously in different bits and parts of the work, everyone is speaking the same language and focused on the same targets and goals, and it all fits together.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*28sN2iWpglfqv13iZHzMLg.jpeg" /><figcaption>Writing your Mission is a huge challenge, it is what you always go back to when things seem out of focus.</figcaption></figure><p>The result can be something really incredible and you end up wanting to hire everyone that did a great job to continue working for you, simply because going through the trouble of explaining everything again for someone else just seems a waste of time. We did our job, but we were also lucky we found truly amazing people that decided to take our challenge as their own. So, our special thanks to everyone that left everything on that table with us: we will always remember your work ethics and dedication. That intensity just bonds people, and it is heartwarming to see them sharing the same dreams ❤</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*A26C7r8YNk6Fzde0V1XfkQ.jpeg" /><figcaption>At a Mesa, you not only need a talented team, but also a committed one! We were more than lucky to get that!</figcaption></figure><p>We didn’t have the biggest budget, so we had to make some tough choices: usually there are 12–14 people at a Mesa plus the leaders. We only had 7 people, our team being 3 of those, which meant being leader, maker, decision-maker and everything else you need. <strong>Being the Mesa leader and a maker at the same time was a huge challenge and I only recommend it if you have a good relationship with the people you are working with, because there might be a lot of operational challenges along the way.</strong> My biggest mistake was not hiring someone to work on sensitive issues, such as space, food and comfort. Worrying that everyone was fed as well as productive was a lot, and I’d go back and just have someone do it for us. <strong>Trust me, it will be a worthy investment.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zhSIg3CGGYqHE5WoOAJZTw.jpeg" /><figcaption>Leading a Mesa can be messy and is VERY challenging, but I’m pretty proud of our achievements! This is a look at Day 2!</figcaption></figure><p>And I believe that is where things could have gone really bad, but we managed to work it out: <strong>our main designer, Bruno, was tasked with the challenge to both coordinate other designers and develop a big part of the website. </strong>But his part was crucial for the development team, and they were depending on it to start programming. I’m really glad <strong>we were able to see that he was overwhelmed in time to get things done.</strong></p><p>On Day 4 his part was not decided yet (which is very late for a Mesa!), so <strong>we called in reinforcements so he could share his ideas with an outsider, helping him focus and clear his mind.</strong> It worked, I wish I could have thought about that sooner, but it is a great resource and I recommend it as a “get that out of your system” kind of process. Having someone with a fresh look can definitely change your perspective!</p><p><em>I also have to thank my incredible designer and sister-in-law, whom I called 7 in the morning to “please come work for me today”, and and hour later she was there with Redbulls and Monsters for everyone. Love you Mary.</em></p><h3>Mesa is all about rituals and processes</h3><p><strong>and there are two I really love: how they begin and how they end.</strong></p><p>You are not supposed to discuss your Mission or any ideas you have on Day 1, and that is scary as hell, because you already have a limited time to work and you already hired and paid for everyone, so you want to get them working asap. But it is extremely important to follow the rules, and totally worth it.</p><p><strong>Day 1 is all about downloading content from outside speakers, internal alignment and getting to know your team members, and if you are not on the same page with them, the rest of your days won’t engage as easy. So</strong> <strong>selecting the right guest speakers and telling the company’s story is as important as anything else to get things done.</strong></p><p>Another ritual I loved was the final presentation. At the end of Day 5 the method demands you present your results to a selected group of people. And that is a real game-changer, because you definitely would leave a lot of things behind if you didn’t have the commitment to present it. We made sure we invited close friends and partners, but family as well, because you never wanna mess up in front of your family members! Creating that deadline and commitment to people you care is very important and it sure helps getting things done.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jAFH-oe0RQ4nmSchp8sKtw.jpeg" /><figcaption>For me it is harder to present to my family than it is for a bunch of strangers! That is a picture of a very tired, but proud Leticia.</figcaption></figure><h3>So what was the result?</h3><p><strong>Well you can check it for yourself, we are pretty proud of it: </strong><a href="http://odd.studio/"><strong>odd.studio</strong></a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JhS88v1XS0r7ZDzCFKN0kw.png" /><figcaption>Check out our results at odd.studio or follow us here or on Instagram!</figcaption></figure><p><strong>We not only achieved in a week — yes, 5 freaking days — our goal to develop our brand, logo, visual identity, but we also had our website online and running. </strong>We met amazing professionals that became good friends, got an incredible amount of other ideas we did not invest and have to explore throughout the year, and participated in the process of building our own brand. <strong>Choosing to do it ourselves was the best decision. I believe hiring someone else to do it wouldn’t have given us the results we were wishing for with the time we had.</strong></p><p><strong>But there was a lot of work into it too: more than 500 hours of dedication, late hours into discussion, some very vulnerable — but heart opening — experiences, hard choices, truthful one-on-one moments with my business associates, litters of coffee, candy cotton Monsters (which I did not know existed), Redbulls and an insane amount of passion and commitment.</strong></p><p>Plus, you work at least 100 hours before and after a Mesa, not everything gets 100% solved and you have a lot of prepping — it took us another month to get pictures, fix bugs and write this post!</p><h3>Was it worth it? Definitely. Do I believe Mesas can actually solve any problem out there?</h3><p><strong>I still have my doubts, but here is where I think it would work and why:</strong></p><p><strong>This process works really well for anything creative, design, conceptual, and even prototyping — physical or digital. It is mind-blowing what you can achieve with the right team and mission.</strong></p><p>So, getting that idea off the paper? Easy. It works for creative processes and data science.</p><p><strong>But things that require exploring and deep-diving I’m still curious to try out. For example, if you have a database you are not familiar with and needs to analyse and explore, maybe this is not the ideal process to start with.</strong></p><p>But building a dashboard once you already know the data? Perfect.</p><p>Trying to analyse and select the best variables for you model, optimizing it to its fullest potential? Tricky, it might get you nowhere, because in data science sometimes you hit dead ends.</p><p>But creating indexes and applying different modeling techniques to a dataset you already know and have can be a great approach. Plus, you’ll get both the data science and the business/area of expertise collaborators in the same room to achieve quick results that otherwise would take endless meetings you to get mildly done.</p><h4>I’m simply in love with the method, I think it is an incredible way of solving problems and I can’t wait to put our learnings into practice for a next challenge, either in our company or with a client’s problem, proving all you really need is good talent and a MESA to achieve greatness.</h4><p><em>ps.: this was a private effort, with no contact whatsoever with the Mesa team, and we are not “officially certified” in any way, so this is my own, personal opinion, and I do not speak for the Mesa team, I only admire what they have created .</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c7695351b169" width="1" height="1" alt=""><hr><p><a href="https://medium.com/weareodd/creating-odd-studio-applying-the-mesa-method-to-build-our-own-brand-c7695351b169">Creating Odd.studio: Applying the Mesa Method to build our own brand</a> was originally published in <a href="https://medium.com/weareodd">oddstudio</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Sobre modelos preditivos e maturidade analítica]]></title>
            <link>https://medium.com/@leticiapozza/sobre-modelos-preditivos-e-maturidade-anal%C3%ADtica-a54b215ca3ed?source=rss-c76f7d1baba8------2</link>
            <guid isPermaLink="false">https://medium.com/p/a54b215ca3ed</guid>
            <category><![CDATA[data-driven]]></category>
            <category><![CDATA[algorithmic-design]]></category>
            <category><![CDATA[predictive-modeling]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Letícia Ange Pozza]]></dc:creator>
            <pubDate>Sun, 17 Sep 2017 15:34:23 GMT</pubDate>
            <atom:updated>2017-09-27T17:19:28.059Z</atom:updated>
            <content:encoded><![CDATA[<p>Sempre que iniciamos um novo projeto na <a href="http://www.cappra.com.br/">Cappra</a> , existe uma rodada que exercitamos a criatividade e a co-criação para que todos possam levantar seus desejos em produtos de dados (dashboards de visualização, relatórios de dados específicos, modelos preditivos, entre outros). Na sequência, reunimos o grupo para priorizar o que será feito primeiro — e sempre surge aquele grupo que quer “fazer um modelo preditivo que vai responder todas as nossas perguntas e melhorar nossa performance/resultado/tomada de decisão”.</p><blockquote>Porém, <em>poucas pessoas sabem que um modelo ou sistema preditivo de dados é tanto uma questão matemática quanto uma questão humana: é, acima de tudo, um processo evolutivo de diversas regras, determinadas pelo negócio ao qual ele atende, e por uma equipe de cientistas de dados, apoiados por uma equipe de tecnologia. Se tivermos que traduzi-lo em uma imagem, seria assim: quanto menor a complexidade, mais próximo de uma descoberta e uma análise aquele dado está. Quanto mais complexa, mais ela incrementa conhecimento, permitindo a criação de regras que alimentam sua evolução e maturidade analítica.</em></blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zHAoG6llWMVhOdlHjQm7wQ.png" /></figure><p><strong>Quê? Eu explico. </strong><em>Um sistema preditivo é uma máquina programada e ensinada a replicar e (constantemente) reaprender a partir de inputs (respostas) que um negócio busca descobrir. Ele apoia um processo de decisão, trazendo a melhor saída ou a mais provável, considerando o histórico de vezes que aquilo já se repetiu, e as diversas variáveis que possam impactar no seu resultado.</em></p><p>Por exemplo: um modelo preditivo pode responder se a demanda de um produto vai aumentar ou reduzir no próximo mês. Para isso, ele aprende com o histórico daqueles dados de demanda e de outras variáveis que possam vir a impactar na demanda, reaprendendo, a cada resposta (valor da demanda real <em>versus</em> o valor que ele previu), melhorando seu resultado de: cresce ou diminui.</p><blockquote>Mas até chegar nessa solução de forma satisfatória (com altos índices de acerto), existe uma maturidade analítica para a qual poucas empresas estão preparadas, pois poucas sabem afirmar, com base em dados, por exemplo, se aqueles valores são sazonais, ou se são impactados pelo dólar, ou ainda, se sobem por causa de datas comemorativas. Geralmente, essas determinações são muito mais na base do feeling do que da prova real.</blockquote><p><strong>E isso acontece pois não se sabe sobre o papel a ser desempenhado pelas partes. </strong>Para construir o modelo, é papel da equipe de cientistas de dados auxiliar na geração de insights sobre os dados e sobre o método, utilizando os dados disponíveis para construir produtos que respondam questões de negócios, com o uso da matemática, da estatística e da visualização da informação. A equipe de negócios, por sua vez, deve determinar quais as decisões que aqueles dados vão responder, e definir as regras que serão aplicadas a eles. Isso se chama <em>problem framing</em> [enquadramento/definição do problema]. Já as regras, são baseadas na constante observação dos dados, no conhecimento que se constrói das informações a partir desses dados, e em objetivos específicos de negócio.</p><blockquote>Se esses objetivos não estão claros, a resposta que os dados do modelo retornarão não será clara. É como ter uma resposta para a qual não temos a pergunta: pode até ser legal, mas é pouco útil. Já se os dados disponíveis não respondem a esses objetivos, a matemática e a estatística não encontrarão respostas, e o modelo não vai trazer resultados satisfatórios.</blockquote><blockquote>É como ensinar alguém a escrever, mas só temos o alfabeto, não sabemos a língua falada ou a série de regras gramaticais que determinam o que vem primeiro, o pronome ou o verbo. Por sua vez, não ter uma base de dados é como se nunca tivéssemos falado uma palavra. E ter uma base de dados ruim é ter cada pessoa escrevendo a mesma palavra de uma forma diferente, não sendo possível chegarmos ao mesmo entendimento ou compará-las entre si. Muitas vezes, a linguagem está ali, mas nós precisamos compreender o que ela significa, e encontrar as regras que foram criadas, pois poucos a conhecem e não há um livro gramatical para nos guiar.</blockquote><p>Quando isso acontece (e isso acontece com muita frequência), é o papel da equipe de cientistas de dados auxiliar a analisar e traduzir essas regras, transformando os dados em informação, e permitindo que a equipe de negócios crie o livro-guia (conhecimento) que possui o mesmo significado e o mesmo entendimento a todos envolvidos. Neste caso, é entender que ter uma grande quantidade de dados, não é sinônimo de valor: o valor está no processo de transformação e não no dado em si.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CwA_S7T43XCLegp__PmPsw.png" /></figure><blockquote>E que, acima de tudo, tenha valor e respostas a questões de negócios.</blockquote><p>Da mesma forma, é o papel da equipe de negócios criar o ambiente propício que permita a análise, a exploração e o questionamento, se engajando para gerar descobertas e testes no dia a dia, não esporadicamente. Afinal, a tomada de decisão está na mão dessa equipe, e não da de cientistas de dados. Desenvolvendo, assim, pensamento analítico crítico e cultura analítica (a caminho da maturidade, lembra?).</p><p>Muitas vezes, o modelo analítico que foi idealizado no início, não é o melhor para os dados que temos disponíveis, ou para o conhecimento que a equipe de negócios tem dos próprios dados. Isso significa mudar a abordagem para um modelo que seja adequado ao nosso contexto e continuar analisando os dados até aprender as regras para construir as respostas que buscamos. E, às vezes, não vamos encontrar as respostas que procuramos nos dados que temos. Isso significa que temos que melhorar e incrementar a base, através de maior tempo de captação, maior comprometimento ao preenchimento, ou até outras fontes.</p><blockquote>Análises preditivas são processos evolutivos: não são simples, nem sempre são lineares, e às vezes podem ser de um processo longo, de idas e vindas, que nos fazem acreditar que não é possível ou que o problema é de quem está calculando os dados. Mas, isso é ciência: desenvolver hipóteses, testá-las, aprender com seus resultados, melhorar e tentar de novo. E por mais que a ciência seja baseada na matemática, ela é guiada pelos objetivos e motivações das pessoas. Saber nosso papel nesse processo é a chave para a boa ciência. E exercê-lo de forma dedicada é o que garante maturidade analítica. E a chave para o caminho a modelos preditivos com resultados excelentes.</blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a54b215ca3ed" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Data Master of None]]></title>
            <link>https://medium.com/@leticiapozza/the-data-master-of-none-d105b0b9a61e?source=rss-c76f7d1baba8------2</link>
            <guid isPermaLink="false">https://medium.com/p/d105b0b9a61e</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[storytelling]]></category>
            <category><![CDATA[analytical-culture]]></category>
            <category><![CDATA[data-scientist]]></category>
            <dc:creator><![CDATA[Letícia Ange Pozza]]></dc:creator>
            <pubDate>Wed, 31 May 2017 16:07:06 GMT</pubDate>
            <atom:updated>2017-05-31T21:06:58.236Z</atom:updated>
            <content:encoded><![CDATA[<p>I&#39;ve read A LOT of articles discussing the required skills and knowledge to become a data scientist. Around the world and in Brazil (where I live), we see tons of new courses every week. And why wouldn&#39;t that be? We&#39;ve heard for the past 5 years that it is the sexiest job on Earth and that we have more allegedly positions than candidates. And those courses address very important issues: from crunching big sets of data to making them predict the future. All technical and essential. Without mapping, mining and visualizing your information, you&#39;re not a data scientist. Or are you?</p><p>What lies beneath that statement is the lack of ability to understand that we&#39;re still talking about organizations. Data scientists per se are not able, by themselves, to organize, manage and create an entire organization. They are part of one. And as any new cell in a pre-set environment, if the company&#39;s culture is not open to questioning, failing and rearranging what they thought were their beliefs, they can have the best technical data scientists in the world: the corporation will eat them up, and they will disappear before they even start their jobs.</p><p>And that happens for a simple reason: data scientists aren&#39;t the ones making the decisions at the end of the day. If companies don’t hire people to help understand this culture, making the right questions and translating the value of the data to the ones deciding where the company is headed, they will keep making the wrong choices. Or what&#39;s even more common: ask questions they already know the answers to. No matter how big or how small the company is, we are still talking about at least 10 solid years of people making big decisions based on feelings, not numbers.</p><p>I&#39;m what they call a creative data scientist. I&#39;m specialised in social non-structured data (creating structured models to analyze big sets of all kinds of social data — from conversations online and offline, to performance and demographics on social media). I also take great pleasure at understanding a company&#39;s analytical need and creating amazing data visualizations models that help them on a daily basis. That&#39;s the &quot;type of data scientist&quot; that I like to be, and there are tons of technical courses for my area of expertise.</p><p>But for the past year I&#39;ve been helping my team of data scientists at Cappra Data Science build better analytical models for all kinds of organizations — from clustering models for marketing or human resources, to predictive models for trade and sales. We are inside retail, commodities, advertising, communications, common goods, tech, big, small, private, governmental, non-profitable…all sorts of companies.</p><blockquote>And when I&#39;m asked what&#39;s the most important skill to help my team nowadays, I always get surprised faces when I say &quot;storytelling&quot;. Storytelling done right: turning complex into simple, clear stories, paired with context comprehension and data science problem-solving, becomes the biggest weapon against narrow-minded or culturally closed decision-makers.</blockquote><p>And that&#39;s not their first worry when talking about learning or teaching how to be a data scientist. That&#39;s why I had to become a Data Master of None: I understand all the basic concepts of data science. Some more, some less. But I have to know clearly:</p><p>[1] what&#39;s the problem we are solving;</p><p>[2] which of our team members can get us the best technical solution;</p><p>[3] who are the stakeholders we are assessing and who will make use of that solution on a daily basis;</p><p>[4] what&#39;s the analytical maturity of that company — are they used to making decisions based on feelings or do they use qualified, organized data to do so? (I&#39;ll talk more about that on my next post);</p><p>[5] but most importantly: how will I tell them why we&#39;ve come up with that solution, how the company benefits from it, and how their team can implement it so they can start using it.</p><blockquote>If we&#39;re not able to show the value of what we create in a simple, logical and applicable way, the biggest data science solution becomes obsolete before it has even started. Not because they are not applicable, but because data science is a match between data problem-solving, technical complexity and people. And more often than not, the first one is not weighed against the others.</blockquote><p>And usually, that doesn&#39;t happen overnight: it takes time, patience, commitment and understanding the limits of what you can accomplish with that company, in their social, economical, hierarchical moment. It&#39;s cheesy, but in some cases, you have to start small, learn about the business, not just about the data, and build up. Plus, it is very unlikely for you to find these set of skills in one data scientist, we&#39;ve learned to shift and combine people&#39;s skills in our projects to deliver the best we can.</p><p>From what I see, with the low level of “analytical maturity”, and the lack of understanding of the job of a real data scientist in companies, we need less technical data scientists and more “data master’s of none” (patent pending ;). Only then, we will be able to start planting the analytical culture seed in companies, and really see the transformation that data science can deliver. We will only be all technical data scientists when we understand that competences related to storytelling, context comprehension and business understanding, are also technical.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d105b0b9a61e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[APRENDIZADOS DE UM CABELO ROSA]]></title>
            <link>https://medium.com/@leticiapozza/aprendizados-de-um-cabelo-rosa-63610127e96b?source=rss-c76f7d1baba8------2</link>
            <guid isPermaLink="false">https://medium.com/p/63610127e96b</guid>
            <category><![CDATA[machismo]]></category>
            <category><![CDATA[negócios]]></category>
            <category><![CDATA[autoconhecimento]]></category>
            <category><![CDATA[feminismo]]></category>
            <dc:creator><![CDATA[Letícia Ange Pozza]]></dc:creator>
            <pubDate>Wed, 07 Sep 2016 01:06:04 GMT</pubDate>
            <atom:updated>2016-09-07T16:09:40.499Z</atom:updated>
            <content:encoded><![CDATA[<p>06.09.2016</p><h3>Sempre quis ter cabelo rosa — tento pintar o cabelo de cores consideradas “diferentes” desde que a minha primeira mesada permitiu isso. Eu tinha 12 anos e queria pintar as pontas do cabelo de azul. Afinal, se algo desse errado, era só cortar as pontas fora. A tentativa não deu certo: as pontas ficaram cinzas e o azul não pegou. Mais tarde, descobri os segredos do descolorante para o meu cabelo, e foi só alegria: aos 13 pintei de vermelho, azul e roxo. Isso tudo, obviamente, escondido e manchando a parede do banheiro de papel crepom, deixando minha mãe bem maluca.</h3><p>Ao longo dos anos (e se for pensar exatamente quando isso aconteceu eu não consigo me lembrar), fui criando um senso de que se eu quisesse ser uma boa profissional, eu deveria ter cabelo das cores tradicionais: castanho, moreno ou loiro. Afinal, ruivo iria parecer <em>falso demais</em> e não me cairia bem. Além dos namorados babacas que não me deixavam ter cabelo loiro pois <em>“eu chamaria atenção na rua e os outros não podem ficar olhando o que é meu”</em>, tinha o mundo profissional que dizia que <em>“eu era bonitinha demais pra ser levada a sério, imagina se eu pintasse o cabelo de cor diferente”.</em></p><p>Como comecei a trabalhar e estudar cedo, eu fui adotando esse ar “corporativo”, como se ele fosse uma entidade suprema ou uma religião que todas as pessoas acabam aderindo em algum ponto da vida. Fui ensinada na escola, na faculdade e no trabalho que me vestir, me portar e falar de forma “corporativa”(olha ela aí de novo) era a melhor forma para que eu me destacasse como uma profissional, e, consequentemente, fosse levada a sério.</p><p>E por muito tempo eu acreditei que funcionava, que era isso que me fazia ter um bom currículo e uma boa reputação de mercado. Posso dizer que me comportar de determinada maneira, aprender a falar em público e com pessoas de níveis hierárquicos muito altos de forma eloqüente e com os jargões necessários, foram competências necessárias para que eu me desse bem e me destacasse em entrevistas e reuniões de negócios. Ou ao menos, foi que sempre acreditei: 1) se vista bem; 2) aprenda sobre o universo daquela pessoa e incorpore ele; 3) quando a oportunidade surgir e você ganhar confiança, mostre quem você é através do seu trabalho; 4) repita 1 e 2.</p><h3>Era quase como se eu vestisse uma máscara para mostrar que eu era capaz de incorporar aquelas convenções sociais e ainda ser uma boa profissional. Era eternamente um trabalho duplo, um <em>modus operandi</em> que me permitia aplicar os conhecimentos de negócios que eu havia estudado. Mas, definitivamente, não refletia o que eu pensava ou como eu me sentia: no fundo eu me sentia falsa, repreendida e sem identidade própria. Um eterno conflito entre quem sou <em>versus</em> o que eu quero ser.</h3><p>E o mais engraçado é que, ainda passando por esse processo todos os dias, eu ouvi coisas como:</p><blockquote>“Letícia, usando essa calça você não é feminina o suficiente”; “você parece frágil com essas suas unhas rosas”; “você desafia a autoridade demais para sua idade, onde já se viu questionar a solução que estamos passando para o cliente?”; “mas você é a mulher que mais ganha dentro desse departamento, isso não é o suficiente?”; “cuidado para não falar sobre sexo dessa forma, você pode ser mal interpretada”; “você é criativa demais para essa empresa, cuidado para não ser demitida por dar ideias demais”, “legal esse vestido colorido, mas não te esqueças que eu te contratei pois tu pareces uma pessoa respeitável e de negócios”.</blockquote><p>Não há necessidade nenhuma em dizer que sempre tive chefes homens, ou mulheres, que tentavam impressionar outros chefes homens.</p><h3>Então, eu me dei conta de que por mais que eu usasse essa identidade, eu sempre seria julgada pela minha aparência e não pelas minhas competências.</h3><h3>Independente do quão corporativa eu fosse, o problema daquelas pessoas não era a minha aparência, eram questões muito maiores que isso. Eu estava sendo julgada e discriminada pelo meu gênero e não só pela forma com que eu me vestia ou pela cor do meu cabelo. Que independente do meu cargo, da minha posição, do status, do conhecimento ou poder aquisitivo, minhas capacidades sempre seriam avaliadas em segundo plano. Corporativa ou não, o meu esforço para convencer aquelas pessoas de que eu era uma boa profissional e de que sou, sim, valiosa para a equipe, seria sempre o mesmo.</h3><p>E, nesse momento, eu me dei conta de que eu estava olhando para isso de forma errada. Fodam-se roupas, foda-se cor de cabelo, foda-se postura ou linguagem: eu podia ser o que eu quisesse. E, livre da sensação de ser julgada pelas roupas, pela forma como me expresso e pela cor do meu cabelo, eu finalmente pintei o cabelo de rosa. E pela primeira vez, eu me senti livre, aos 27 anos, para ser quem eu sempre quis ser, da cor que eu queria ter. E pra mim foi incrível, uma experiência transformadora, fisicamente e emocionalmente.</p><h3>Mas o que eu não imaginava é que, com o cabelo rosa, eu finalmente daria um motivo explícito para as pessoas demonstrarem o machismo delas em relação mim. Hoje, não falam mais “a menina bonitinha demais”, “novinha demais”, “pretensiosa demais pra idade dela”, “questionadora, curiosa ou criativa demais”, questões que não eram feedbacks ou algo no que eu pudesse trabalhar como competência. Finalmente, é “não serve, pois ela tem cabelo rosa”.</h3><h3>E isso ainda não me define como profissional, mas ao menos explicita a incapacidade daquela pessoa de perceber meu potencial como profissional. E me deixa claro que, com aquela pessoa, eu não quero e não preciso trabalhar, pois ela não merece meu respeito, esforço e meu trabalho.</h3><p>Não vou dizer que ainda não dói. Dói demais e ainda estou aprendendo a lidar com o processo. Mas é de um empoderamento indescritível. Eu finalmente sei. Sei por que não me querem numa palestra e preferem que outro homem menos preparado ocupe meu lugar. Sei por que as pessoas me olham na rua (e velhinhos encaram e balançam a cabeça negativamente). Sei por que eu sou confundida como aluna e não professora (e por que riem da minha cara ou fazem piada de mal gosto quando digo que dou aula).</p><h3>É um saber muito reconfortante. Pois me permite ser e não parecer.</h3><p>Hoje me identifico com o rosa. Mas não sei qual a cor de amanhã ou da próxima semana. Sou muito feliz por representar uma empresa que me entende, me incentiva e me respeita pelo que sou, e não pela cor do meu cabelo. Uma empresa que junto a mim levanta a bandeira dessa imagem que fica tão clara do machismo no mundo corporativo. E espero que essa imagem que fica tão clara pra gente, possa ser refletido de volta para eles quando alguém questionar “mas a gente não quer trabalhar com ela por causa de um cabelo rosa?”. Que esses aprendizado sejam deles, tanto quanto meus.</p><p><em>Nenhum cabelo foi danificado durante a execução dessa postagem.</em></p><p>Obs.: um adendo bem importante que me lembraram de incluir: as crianças. Pintar o cabelo de rosa me fez ser uma sensação entre meninos e meninas com menos de 10 anos. E o mais legal é que deles eu nunca vi questionamento ou julgamento: é uma surpresa genuína e positiva. Me alegra pensar que posso ser um exemplo de uma pessoa mais velha com o cabelo colorido que mostre a eles que podemos ser o que quisermos. Só tenho dó das mães e dos pais depois, que tem que ficar ouvindo &quot;Mãe, eu também quero o cabelo rosa!&quot; :P</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=63610127e96b" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>