<?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 Kudratbek on Medium]]></title>
        <description><![CDATA[Stories by Kudratbek on Medium]]></description>
        <link>https://medium.com/@kudratbek?source=rss-fe576e6ad3b8------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*kywarNWoXtY_K_ePUbkEbA.jpeg</url>
            <title>Stories by Kudratbek on Medium</title>
            <link>https://medium.com/@kudratbek?source=rss-fe576e6ad3b8------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 12:13:48 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@kudratbek/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[Where to apply for a Job in South Korea as a foreigner]]></title>
            <link>https://medium.com/@kudratbek/where-to-apply-for-a-job-in-south-korea-as-a-foreigner-a34a36732d76?source=rss-fe576e6ad3b8------2</link>
            <guid isPermaLink="false">https://medium.com/p/a34a36732d76</guid>
            <category><![CDATA[job-hunting]]></category>
            <category><![CDATA[korea]]></category>
            <category><![CDATA[job-search]]></category>
            <category><![CDATA[korean-jobs]]></category>
            <dc:creator><![CDATA[Kudratbek]]></dc:creator>
            <pubDate>Wed, 30 Apr 2025 10:16:38 GMT</pubDate>
            <atom:updated>2025-04-30T10:16:38.736Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/398/1*Fa8OizHeg0PQ8qeJtGpaPg.png" /></figure><p>Are you in South Korea who studies here or wants to move here and want to land job here? What is the method of landing your job interview and land your first job, you are covered with this article.</p><p>First of all, roadmap depends on your korean language skills and your industry skill expertise. Being professional, who is really good at what you are doing, is gonna play a huge role in this process. Initially, lets go through the process for person with good korean skills. You must be good at communicating in korean with koreans, if you have study/studied in korea, you must have amazing communicating korean and korean culture.</p><p>Let’s look at platforms that you can apply for jobs. Korean job appyling platforms are quite different than global ones. Let’s go through with popularity by job postings in order.</p><p><a href="https://www.saramin.co.kr/zf_user/">Saramin</a> job searching platform which is the most popular for job posters and searchers. <br><a href="https://www.jobkorea.co.kr/">JobKorea</a> is also another popular one.<br><a href="https://www.wanted.co.kr/">Wanted</a>, <a href="https://career.rememberapp.co.kr/job/postings">Remember</a>, <a href="https://jumpit.saramin.co.kr/">Jumpit</a> are more orianted for Tech related jobs. <br><a href="https://www.linkedin.com/feed/">LinkedIn</a> is popular Globally but not that much in South Korea.</p><p>As a person who knows korean and know the profession, you should follow the order to get most chance to get job in korean company. It’s better to write your Resume in korean and write self introduction(자기소개서) in one of these local job hunting platforms. Why? In most cases, HR or someone in management who might do not know English will be seeing your application. It’s just discomfort for them and hints that you have poor or no korean skills if the application is written in English. That leads to avoiding to deal with even. Though you are foreigner, write almost everything in Korean. They are most likely to call you for in-person job interview. You decide best possible day and time. Getting on-site interview means 50% job done, now only 50% is done. Do your homework about company and position that you are applying. Be confident and land your JOB. Good Luck!</p><p>Now for a person who is bad to no korean skills should follow the list below to that gives highest chance in order:</p><p><a href="https://www.linkedin.com/feed/">LinkedIn</a><br><a href="https://www.wanted.co.kr/">Wanted</a>, <a href="https://career.rememberapp.co.kr/job/postings">Remember</a>, <a href="https://dev-korea.com/">Dev-Korea</a> (mostly tech jobs)<br><a href="https://www.saramin.co.kr/zf_user/">Saramin</a>, <a href="https://www.jobkorea.co.kr/">JobKorea</a><br>Look for jobs that has description in English where the job opening is going to be more International orianted jobs where company does business or plan to do so with other countries. Focus on hard skills that gives edge for that job that you might be much highly skilled who is better at job comparing to others who speaks Korean.</p><p>Bonus tip is apply more jobs, do more research. Sometimes quantity is much crucial than quality. Quality is must have but many peaple focus on quality and get frasturated so give up which is the worst outcome. Reaching out or being showed to as many companies as possible where you might be only or best candidate who applied to that particular company. Sometimes you just get lucky by applying a company that recently posted job, be there to apply to where they need employee urgently. That’s it, you get kick started your career in South Korea.</p><p>#Korea #jobsearching #resume #chaebol #linkedin #jobhunting</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a34a36732d76" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Finding Your Match in Machine Learning: A Beginner’s Guide to Choosing Algorithms]]></title>
            <link>https://medium.com/@kudratbek/which-ml-algorithms-are-right-for-you-in-data-science-and-data-analysis-66aeadea6c0f?source=rss-fe576e6ad3b8------2</link>
            <guid isPermaLink="false">https://medium.com/p/66aeadea6c0f</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[computer-science]]></category>
            <category><![CDATA[algorithms]]></category>
            <dc:creator><![CDATA[Kudratbek]]></dc:creator>
            <pubDate>Thu, 25 Jan 2024 05:49:16 GMT</pubDate>
            <atom:updated>2024-01-25T05:58:56.692Z</atom:updated>
            <content:encoded><![CDATA[<p>So you want to do a project for either work or study purposes. When thinking about which algorithm is going to be the best fit for the Machine Learning(ML) model that you are thinking about doing. Things you need to consider are:</p><ol><li>The size, quality, and data type</li><li>Computational time and urgency of the task in hand</li><li>What you plan to do with the data</li></ol><p>It is impossible to tell the best-fit algorithm for all cases, even professional data scientists can’t tell upfront. But there can be some most likely algorithm suggestions. Now let’s take a look ML algorithm types.</p><h3>Supervised learning</h3><p>Supervised learning algorithms are used to make predictions about the future based on labeled past data. The algorithms that fall into this category are below:</p><blockquote>Classification: Data is used to predict dog or cat based on given labeled or instructed data on an image. Classifying two things are binary classification and tree or more are called multi-class classification.</blockquote><blockquote>Regression: Predicting continous values within the givin data</blockquote><blockquote>Forecasting: Making predictions based on past and present data to understand trends, mostly used in sales.</blockquote><h3>Semi-supervised learning</h3><p>As mentioned above, supervised learning needs labeled data. But sometimes getting labeled data can be impossible due to high cost and time. In this case, semi-supervised learning comes in handy to make our models better and helps us to get more accuracy.</p><h3>Unsupervised Learning</h3><p>Using fully unlabeled data gives an extra edge to unsupervised learning algorithms over the others. This algorithm detects patterns within the given data. However, it can lead to lower accuracy compared to supervised learning.</p><blockquote>Clustering: Making groups of similar data that makes easy to do in depth research within the given dataset.</blockquote><blockquote>Dimension reduction: Some datasets contain very noise data featers that might distract you. Reducing dimensionality is key for finding true relationship</blockquote><h3>Reinforcement learning</h3><p>No need for any data for training the model, instead an agent interacts with the surrounding environment by doing one action and receiving reward or punishment from the environment/user. The agent decides the next action based on the feedback it got for its previous actions. In recent years it has been used widely in games, robotics, and control.</p><h3><strong>Choosing the right algorithm …..</strong></h3><p>When starting with machine learning (ML), it’s crucial to select the right algorithm based on the problem you’re solving, the data you’re working with, and the type of outcome you desire. Here’s a simplified guide:</p><p><strong>Linear Regression:</strong> Use this when you have a continuous output variable and you’re looking to predict its value based on the linear relationship with input features.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*_rjB3OGgJzi8O-Y4.png" /><figcaption><strong>Linear regression</strong></figcaption></figure><p><strong>Logistic Regression</strong>: Opt for logistic regression when your output variable is categorical (e.g., yes/no, true/false). It predicts the probability of the output belonging to a particular class.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*PA9K9WOyjsxfLWr5.png" /><figcaption>Logistic regression</figcaption></figure><p><strong>Support Vector Machine (SVM)</strong>: This is suitable for both linear and non-linear classification tasks. If your data is not linearly separable, use the kernel SVM to map your data into a higher-dimensional space where it is linearly separable.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/781/0*SoKGc2JWgkMBPvjd.jpg" /><figcaption><strong>Support Vector Machine (SVM)</strong></figcaption></figure><p><strong>Decision Trees: </strong>These are easy to understand and are useful for classification and regression tasks. They work well when you have categorical input variables.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/358/0*xgQl0L7cOlaNkSVw.png" /><figcaption>Decision tree</figcaption></figure><p><strong>Random Forest and Gradient Boosting:</strong> Both are ensemble methods that use multiple decision trees to produce a more robust and accurate model. They help to prevent overfitting, which can be a problem with individual decision trees.</p><p><strong>Neural Networks:</strong> These are powerful tools for both classification and regression, especially in complex problems like image and speech recognition. Deep neural networks, which contain many layers, can capture complex patterns in data.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/444/1*LSoxhbpagy9PLEg4wpOWWA.png" /><figcaption><strong>Neural Networks</strong></figcaption></figure><p><strong>k-means/k-modes: </strong>These algorithms are best for clustering tasks where you want to group your data into k distinct clusters. k-means is used for hard clustering (each data point belongs to exactly one cluster), while k-modes is suited for categorical data.</p><p><strong>Gaussian Mixture Model (GMM): </strong>GMM allows for soft clustering, where each data point can belong to multiple clusters with different probabilities.</p><p><strong>DBSCAN: </strong>This algorithm is ideal when you do not know the number of clusters in advance. It groups data points into clusters based on their density.</p><p><strong>Hierarchical Clustering:</strong> Use this when you want to create a hierarchy of clusters and you do not necessarily need to specify the number of clusters in advance.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/979/0*8LUpbwJfHbnrw2Wn.png" /></figure><h3>In conclusion</h3><p>Here’s a streamlined workflow for tackling new ML problems:</p><p>1. Problem Identification: Clarify the issue you’re addressing.<br>2. Begin with Basics: Get to know your data and establish initial results.<br>3. Advance Gradually: Progress to more complex methods as needed.</p><p>In line with this approach, it’s beneficial to start with straightforward algorithms like linear or logistic regression to get a foundational understanding. As you become more comfortable, you can explore sophisticated techniques such as ensemble methods or neural networks to enhance your model’s performance while increasing model accuracy in parallel to your skills of working with ML.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=66aeadea6c0f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[This is my first medium article]]></title>
            <link>https://medium.com/@kudratbek/this-is-my-first-medium-article-65592f7a38bd?source=rss-fe576e6ad3b8------2</link>
            <guid isPermaLink="false">https://medium.com/p/65592f7a38bd</guid>
            <category><![CDATA[machine-learning-ai]]></category>
            <dc:creator><![CDATA[Kudratbek]]></dc:creator>
            <pubDate>Mon, 17 Jul 2023 12:40:43 GMT</pubDate>
            <atom:updated>2023-07-17T12:40:43.942Z</atom:updated>
            <content:encoded><![CDATA[<p>This is my first medium article</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=65592f7a38bd" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>