<?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 Bruno Paixão on Medium]]></title>
        <description><![CDATA[Stories by Bruno Paixão on Medium]]></description>
        <link>https://medium.com/@brunopaixo?source=rss-a2a6b0beb6c5------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*7sBX2J74EzDHtuPz.</url>
            <title>Stories by Bruno Paixão on Medium</title>
            <link>https://medium.com/@brunopaixo?source=rss-a2a6b0beb6c5------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 02:21:15 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@brunopaixo/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[Using Data Analysis and Machine Learning to Identify Violence Zones in Somalia]]></title>
            <link>https://medium.com/omdena/using-data-analysis-and-machine-learning-to-identify-violence-zones-d08861ebd51c?source=rss-a2a6b0beb6c5------2</link>
            <guid isPermaLink="false">https://medium.com/p/d08861ebd51c</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[climate-change]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Bruno Paixão]]></dc:creator>
            <pubDate>Thu, 24 Oct 2019 00:35:56 GMT</pubDate>
            <atom:updated>2020-01-05T00:34:30.798Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/843/1*RpObDKVqblZlSwZnT7H6Ig.jpeg" /><figcaption>Forced displacement in Somalia</figcaption></figure><h4>SVM classifiers with an average accuracy of 99% to monitor critical areas of conflict including the type of incidents and impact.</h4><p>The conflicts in Somalia have reached alarming levels, year after year many people are victimized by disputes of territory and dominance of spaces. The problem has reached intolerable levels in the international community.</p><p>This report aims to inform intervention actions through insights that are placed as strategic tools for facing the presented problems. The work is part of Omdena’s AI challenge in partnership with the <a href="https://www.unhcr.org/">UNHCR — The UN Refugee Agency</a>.</p><blockquote>The data is derived from a wide variety of local, regional and national sources and the information is collected by trained data experts around the world.</blockquote><h3><strong>The data set</strong></h3><p>The first step in discovering knowledge is to perform data exploration. Many data scientists despise the power of a good data visualization. For this step, I used pandas profiling. The library optimizes the process of variable analysis, providing information that requires development time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/457/0*jcji5at-Di4Z3Kls" /><figcaption>Dataset</figcaption></figure><p>A simple line of code provides a real picture of the variables and this is critical when building the data model. The dataset has 31 variables, 10 numerical, 14 categorical, 6 rejected, the model already identifies and finally a text field variable. Thus we can select which variables the data model will compose of.</p><h3><strong>Data Insights</strong></h3><p>Building good visuals follows an unbeatable formula: less is more. In this sense, graphics and plots need to be as synthetic as possible, delivering a value chain for those who need to make a decision based on the data presented.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*3PrLPtJXfEGFRe-i" /><figcaption>Graph 1 — Total Cases according to type over the years</figcaption></figure><p>This graph shows the evolution of cases over the years.</p><h4>Insights</h4><p>It is possible to observe that “battles” always had 50% of incident cases, but from 2017 this type lost space in proportion. Directly proportional to the reduction in battles has been an increase in the “Explosion / Remote Violence” category, indicating that new technologies such as drones or remote explosions are gaining ground in violent conflict practices. Finally, in this chart, we can see a consistent increase in violence against civilians over the years.</p><p>In the following graph we can see the lethality level by type of “sub-event”, it is evident that in absolute numbers “armed clash” is what has lethality, followed by “attack” and in third place “air/drone”. These data indicate a trend in case types that most impacts on the number of fatalities.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/752/0*nf4gmT7F_xbwXU7r" /><figcaption>Graph 2 — Fatality Level by Sub Event</figcaption></figure><p>Evaluating relative data has great power in decision-making data evaluation, in Graph 3 we can see how a variable aggregation measure: total cases by ”actor1” and average fatalities provides a panel of performance for decision making. Divided by colors and signs we can easily see the serious cases that need immediate intervention.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/558/0*lXEX_FsC9xZev2OX" /><figcaption>Graph 3 — Actor Lethality Monitor</figcaption></figure><p>Data visualization has a fundamental and decisive power in structuring and generating insights for decision making, exploiting this information brings the user a way to build solutions to confront the presented cases.</p><h3><strong>The SVM tool</strong></h3><blockquote><em>SupportVector Machine (SVM) is a supervised algorithm Machine Learning that can be used for classification or regression. Despite the slightly confusing name, you will see that it makes perfect sense. SVM can be translated as Support Vector Machines. Supporting Vectors are the basis of how the algorithm works because their architecture depends on these vectors. — Felipe Santana</em></blockquote><blockquote>Mining data by deriving insights from data visualization, we have the necessary elements for building machine learning models. The purpose of this data model is to build a hot zone solution, which seeks to predict the most dangerous locations and the highest fatalities. <strong>This type of prediction is extremely business-friendly because this model optimizes the leasing of utility personnel to handle the cases presented.</strong></blockquote><p>To build this model it was necessary to create a new feature called “mild_of_fatalities”, this new variable is categorical value labeled “high” and “low”. To establish this new category we used the mean, standard deviation, and variance of the data column of the “fatalities” column. Thus, results from the “fatalities” column with values ​​above six cases were considered as “high”, results below six cases were considered as low. This new variable characterizes the regions between very dangerous and non-dangerous.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/308/0*0jyWN1hU5dN0LTsw" /><figcaption>Figure 1 — Creating a new feature</figcaption></figure><p>From this point, we can establish the target variable of the data models. The next step was to define the predictor variables, for this was applied the technique of “GridSearch”, thanks to this model it was possible to establish the degree of importance of the predictor variables. The following variables were defined: time_precision, event_type, latitude, longitude, location, fatalities.</p><p>Having defined the target variable and the predictor variables, the next step is to perform the data transformation process. In this stage, the “Standard Scale” and “label encoder” were applied to perform data normalization.</p><p>The fact is that SVM only works with numeric data, thus, this algorithm does not accept categorical values. Therefore we need to convert the column “Location” to numeric.</p><p>One Hot Encoder converts categorical values ​​into binary vectors. The result of this technique is a presence matrix where the columns are the categorical data and the rows the presence.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/530/0*c2S22nj8LyNUtvMb" /><figcaption>Figure 2</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/812/0*IY1xJw_nean8E4sa" /><figcaption>Figure 3 — One Hot Encoding</figcaption></figure><p>Let’s now test SVM with this setting, to do this let’s use the pipelines:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/325/0*txHiVo0Pk8pxfI3G" /><figcaption>Figure 4 — Pipeline</figcaption></figure><p>With just LabelEncoder the model was simpler, with fewer features, this was a considerable change in performance. from SVM.</p><h3><strong>The Results</strong></h3><p>We can see that the rbf kernel (pipeline 2) performed better with an average accuracy of 99%. The other kernels underperformed.</p><p>This result can change dramatically if we change the data. Kernels do data transformations, which means that for each situation a kernel can be more interesting.</p><p>Another point to note is that Kernels makes the model more complex. This can also be a problem.</p><p>This gives us a more detailed view of classifier performance, including by class, so you can see how it is doing in classifying specific classes.</p><p>With an average accuracy and the 99% measurement, I believe we have a good SVM rating result.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/565/1*fTeas1XdkiF6h-LxIDq_ew.png" /></figure><h3><strong>Conclusion</strong></h3><p>This model attempts to predict the most dangerous locations. The applications for the program are diverse because, based on the number of deaths, the system seeks to provide the user with the intervention zones of the security forces, as well as the creation of programs and public policies aimed at reducing violence.</p><h4>Want to become an Omdena Collaborator and join one of our tough AI for Good challenges, <a href="https://omdena.com/our-projects/">apply here</a>.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/875/0*dS2ox6em8zL_ykTB.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d08861ebd51c" width="1" height="1" alt=""><hr><p><a href="https://medium.com/omdena/using-data-analysis-and-machine-learning-to-identify-violence-zones-d08861ebd51c">Using Data Analysis and Machine Learning to Identify Violence Zones in Somalia</a> was originally published in <a href="https://medium.com/omdena">Omdena</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How Artificial Intelligence Can Help in Cases of Sexual Violence in India.]]></title>
            <link>https://medium.com/omdena/how-artificial-intelligence-can-help-in-cases-of-violence-in-india-7ded1b0945?source=rss-a2a6b0beb6c5------2</link>
            <guid isPermaLink="false">https://medium.com/p/7ded1b0945</guid>
            <category><![CDATA[violence]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[india]]></category>
            <category><![CDATA[women]]></category>
            <dc:creator><![CDATA[Bruno Paixão]]></dc:creator>
            <pubDate>Tue, 24 Sep 2019 01:53:05 GMT</pubDate>
            <atom:updated>2019-10-10T19:43:31.813Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Using data visualization to bring new insights and potentially help thousands.</strong></p><h3>Problem definition</h3><p>India can be a dangerous country for women, threatened by the high risk of sexual violence and being forced into slave labor. The finding is a survey with 550 specialists in women’s issues held by the Thomson Reuters Foundation. Afghanistan and Syria, countries submerged in war, are second and third-placed in the ranking. Next, are Somalia and Saudi Arabia.</p><p>The news and numbers paint a dramatic portrait of the situation of women in India. Only in the last month, three different episodes involving teenagers who were raped and burned by their tormentors drew attention to the world. In 2015, the country recorded 35000 cases of sexual violence, but it is possible that this number is much higher.</p><p>Rape is the fourth most common crime against women in India. According to the 2013 National Crime Records Bureau (NCRB) annual report, 24,923 rape cases were reported across India in 2012. Of these, 24,470 were committed by someone known to the victim (98% of cases).</p><p>India has been characterized as one of the “countries with the lowest per capita rape rate.” A large number of violations are not reported. The willingness to report rape has increased in recent years after several incidents of rape have received widespread media attention and provoked public protests. This led the Government of India to reform the penal code for the crimes of rape and sexual assault.</p><p>According to NCRB 2015 statistics, Madhya Pradesh has the highest gross number of rapes reported among the states of India, while Jodhpur in Rajasthan has the highest per capita rape rate followed by Delhi, the capital.</p><p>Prior to February 3, 2013, Section 375 of the Indian Penal Code had defined rape as:</p><p>§375. The rape. A man is said to consolidate “rape”, which, except in the following case, except has sex [19] with a woman in circumstances falling into any of the following six descriptions: -</p><p>First of all. — Against your will.</p><p>In second place. — Without your consent.</p><p>In third place. — With your consent, when her consent has been obtained by placing her, or anyone in whom she is interested, in fear of death or pain.</p><p>In fourth place. — With your consent, when the man knows that he is not her husband, and that consent is given because she believes he is another man to whom she is or believes to be legally married.</p><p>Fifth. — With her consent, when, at the time of giving such consent, on grounds of mental insanity or drunkenness, or administration by him personally, or through another of any numbing or unhealthy substance, she is unable to understand the nature and the consequences of what she gives her consent to.</p><p>Sixth. — With or without your consent when she is under sixteen [20] years of age.</p><p>Explanation. — Penetration is sufficient to constitute the sexual relationship necessary for the offense of rape.</p><p>Exception. — Sexual intercourse by a man with his wife, the wife not being at fifteen years of age, is not rape.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*TGijXCHBezvwZ3zS.png" /></figure><h3>A community building AI solutions</h3><p>To solve this problem a group of AI collaborators at Omdena gathered around a purpose: how can artificial intelligence present paths to solve this problem?</p><p>Today we live in the data age and the automated information. Real-time data analysis in addition to increasingly sophisticated system recommendations. In a philanthropic way, several developers met with the objective of presenting clear and objective solutions in coping with this problem. In addition to presenting solutions, these developers, to which I include, structured scalable solutions to solve problems. With only one “. CSV” file, a community has structured itself with the central focus on building solutions to meet this challenge.</p><p>A product that I could actively contribute was the “get me Out”, a panel that provides the closest relief equipment to a crime zone. Perhaps the most important step for a victim is to seek help. Information that is sometimes dispersed and with a low organization to provide an emergency service.</p><p>With this tool, it is possible to find the main risk zones by type of crime. See as an example New Delhi, a region with several cases of “catcalls”.</p><p>Figure 1— New Delhi Cases</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/439/1*-YYJ2CYly49ZOzqGh8-e9Q.png" /></figure><p>In addition to establishing “hot zones” of crimes, the system conducts a geoprocessing seeking in which emergency services exist near the occurrence, measuring the distance in kilometers from the nearest relief site. In the case of the example, we can see that there is a police station at 3.16 kilometers from the occurrence.</p><p>Figure 2 — distance from the relief sites</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/523/1*2gkTD2DSwiVuHzPaJeaV4Q.png" /></figure><p>Complementing the tool provides the peak times of occurrences, in addition, it shows the comments on the crime suffered, this concatenated to the geolocation and the nearby equipment.</p><p>Figure 3— DashBoard</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/567/1*9trpV3GpkdmyN_u9pP5Cag.png" /></figure><h3>Final considerations</h3><p>Thus, this tool presents a set of solutions to perform violence, using data and artificial intelligence. This model identifies the closest safe zones for victims to arrive in case of assault.</p><h3>Personal experience</h3><p>This challenge demonstrated the power of the community. I have seen several developers working on the mission to help victims of sexual violence in India. In addition to being able to contribute to the power of data visualization, I could demonstrate that data visualizations can save lives and protect people.<br>The great lesson I draw from this challenge is that the power of connections yields great results.</p><h4>Want to become an Omdena Collaborator and join one of our tough AI for Good challenges, <a href="https://omdena.com/our-projects/">apply here</a>.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/875/0*u2POr-CrYUO2CtoM.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7ded1b0945" width="1" height="1" alt=""><hr><p><a href="https://medium.com/omdena/how-artificial-intelligence-can-help-in-cases-of-violence-in-india-7ded1b0945">How Artificial Intelligence Can Help in Cases of Sexual Violence in India.</a> was originally published in <a href="https://medium.com/omdena">Omdena</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>