<?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 Weerawut Chaiyasomboon on Medium]]></title>
        <description><![CDATA[Stories by Weerawut Chaiyasomboon on Medium]]></description>
        <link>https://medium.com/@weerawutchaiyasomboon?source=rss-4678d24886ea------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*P3CkMkrR57wZg8cw.</url>
            <title>Stories by Weerawut Chaiyasomboon on Medium</title>
            <link>https://medium.com/@weerawutchaiyasomboon?source=rss-4678d24886ea------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 03:50:08 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@weerawutchaiyasomboon/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[I will survive!? — Data Science with Rapidminer]]></title>
            <link>https://medium.com/@weerawutchaiyasomboon/i-will-survive-data-science-with-rapidminer-d03793ab21fc?source=rss-4678d24886ea------2</link>
            <guid isPermaLink="false">https://medium.com/p/d03793ab21fc</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[rapidminer]]></category>
            <dc:creator><![CDATA[Weerawut Chaiyasomboon]]></dc:creator>
            <pubDate>Mon, 21 Sep 2020 04:10:36 GMT</pubDate>
            <atom:updated>2020-09-21T04:10:36.348Z</atom:updated>
            <content:encoded><![CDATA[<h3>I will survive!? — Data Science with Rapidminer</h3><p>Remark: This is my note which I try to apply my study. Any comments or suggestion please feel free to add, very appreciate.</p><ol><li>The purpose of this note is to predict a person can survive or not? This is classification problem.</li><li>Data I use is the Titanic dataset which you can get from Rapidminer (/Samples/data/)</li><li>First we Retrieve Titanic dataset</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/438/1*BBw129FxlKWEO1TuE3xiSw.png" /></figure><p>4. See at Statistics tab. Found out the data has 1,309 examples and 12 regular attributes. And data has some missing values.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9SCBgP29w79QgwxwEHgDTg.png" /></figure><p>5. I will remove attributes that I think not related with to predict. I will remove Name, Ticket number, Cabin, Port of Embarkation, and Life boat. I use Select Attributes operator.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/883/1*EoBjnSrTLiZIjmhEclr5iQ.png" /></figure><p>So we got only 7 regular attributes below.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Yn0xzaNQ4OY4zAyAQ3qypg.png" /></figure><p>6. For data above we found out that Age has 263 missing values and Passenger Fare has 1 missing values. We will replace them with average of that attribute, using Replace Missing Values operator.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/856/1*7flB24fJcazH_QjhvDBrGA.png" /></figure><p>7. Then we use Set Role operator to set Survived attribute to be label.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/615/1*dBu7lPaXk28c-rxq0mbg3Q.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*R4BzB4WYoecu4vZCtG5WVQ.png" /></figure><p>8. We do some data exploration before go to prepare data for modeling. Found out some interesting as below.</p><p>8.1 Third class passenger trend to be more not Survived than orther two classes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lpUOBhgIi_bTLJm-qCYh6A.png" /></figure><p>8.2 Female trend to be more Survived than Male.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9KZ8s7hzvkwXVDxZj9sBWg.png" /></figure><p>8.3 Person Age around 26 to 53 trend to be not Survived than younger or older Age.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cXMWKG-IUFAuyUGPBrjqbw.png" /></figure><p>8.4 Person with low Passenger Fare trend to be not Survived more than higher Passenger Fare.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gT33N9s9C9cKOpdcyDt6pw.png" /></figure><p>8.5 Person with 2 or below Siblings or Spouses trend to be not Survived more than more number.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KHsuNW1cnJ4ZvZ-GsYTwAg.png" /></figure><p>8.6 Person with lower 3 Parents or Children trend to be not Survived more than more number.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*neVQ-O2esFtvTUF77z2akg.png" /></figure><p>** For item 8.3 to 8.6 I used Discretize by binnings for analysis.</p><p>9. Try to predict by Decision Tree with 10 fold Cross Validation. Got Accuracy 79.45%</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/751/1*AMujMxE8GhFxqof0r2lI8g.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Y1qp8Vs_Fj9Vt1dEKP4wtw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Dj-u6dNAD4iBRG1FYC8lmQ.png" /></figure><p>10. Put Select Subprocesses to select different models.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*HAaUdGMktHZfTE1y6YHSZA.png" /></figure><p>11. Try Naive Bayes with 10 fold Cross Validation. Got Accuracy 76.62%</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*A1GCFE3QxyPOpBxZIRf1lg.png" /></figure><p>12. Try kNN with 10 fold Cross Validation. Got Accuracy 69.75%</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8nh4Gc2RY0ABgxw3sSiclQ.png" /></figure><p>13. Try Random Forrest (100 trees) with 10 fold Cross Validation. Got Accuracy 80.75%</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wYOJul3eWnaj04Q9TwhG7Q.png" /></figure><p>14. Try Neural Network (need to use Norminal to Numerical and Group Model, due to data need to be numerical for Neural Network). Got Accuracy 80.90%</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/294/1*HCyf4jEv6G2x8Frcn1zCeQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7mh300b6whgAesItLrUYgg.png" /></figure><p>15. Use Apply Model on Titanic unlabel data with Neural Network model we trained.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/462/1*JQnYtM5xhEKknDvNCR6zVg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EkwppL76IIWM3Ik-Nj5ezA.png" /></figure><p>Yes, I think that’s it for now. Thank you for reading through the end. See you next time!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d03793ab21fc" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Data Science with Python v Rapidminer — Introduction]]></title>
            <link>https://medium.com/@weerawutchaiyasomboon/data-science-with-python-v-rapidminer-introduction-62a423be04e6?source=rss-4678d24886ea------2</link>
            <guid isPermaLink="false">https://medium.com/p/62a423be04e6</guid>
            <category><![CDATA[rapidminer]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[sklearn]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Weerawut Chaiyasomboon]]></dc:creator>
            <pubDate>Fri, 04 Sep 2020 07:43:00 GMT</pubDate>
            <atom:updated>2020-09-04T07:43:00.265Z</atom:updated>
            <content:encoded><![CDATA[<h3>Data Science with Python v Rapidminer — Introduction</h3><p>Note อันนี้เป็นการเรียนของผม โดยผมจะแสดงขั้นตอนการทำ Project Machine Learning โดยใช้ Python เทียบกับ Rapidminer ผิดพลาดประการใด กรุณาแนะนำนะครับ (อันนี้เป็นบทความแรกในชีวิตเลยครับ!!!)</p><p>หมายเหตุ: ตัวอย่างอันนี้ผมเอามาจากหนังสือ <a href="https://www.amazon.com/gp/product/B07XGF2G87/ref=ppx_yo_dt_b_d_asin_title_o00?ie=UTF8&amp;psc=1">Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems</a> (โดยตัวอย่างคือ เราใช้ data housing.csv เพื่อ Predict median_house_value)</p><ol><li>Load ข้อมูลเข้ามาแล้ว แสดง 5 rows แรก โดย Python</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1014/1*m2Qq6OkmdpPdWlh4Hjx9fA.png" /></figure><ol><li>Load ข้อมูลโดย Rapidminer ใช้ Read CSV operator (set role median_house_value เป็น label)</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/542/1*yvD9jaiDInIIDYPMgXy5iA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Ym2X28D25n2e1hd3S9xPKA.png" /></figure><p>2. Data Exploration by Python</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mQVqYYD6cIbpHW9GuDKpLg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*52Av7U0LbIHSfob_KRSgqA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vUC2XcZLbDJEDsYtRjIj7w.png" /></figure><p>2. Data Exploration by Rapidminer</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XYS3uJIh1rFM4pS2SOzmYw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*plZs4QQ3jrBtX68F_0viUQ.png" /></figure><p>3. Correlation by Python</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TB9vco1-oUtWRyq67aUm1w.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/1*qwstj9CR4iwxvKn2ROdWuQ.png" /></figure><p>3. Correlation with Rapidminer</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/539/1*k1zL09ht3_HAKlF9HcVavA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mxFkSn3wN82pWLsYxqznyw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9LUtftlPqDQgt3sKlEWufw.png" /></figure><p>4. Fill missing values by Python</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*owWYdplp1ACBKR0tdPBBQA.png" /></figure><p>4. Fill missing values by Rapidminer</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/919/1*-wnEjhkdE6KLOiW_6bz7CQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uM0Q6J5H0nJj8Vo_qvCPHA.png" /></figure><p>5. Transform Category(ocean_proximity) to Numerical by Python (OneHotEncoder)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1022/1*fBjkZjGvHzhzb4n4vYh2tA.png" /></figure><p>5. Transform Category(ocean_proximity) to Numerical by Rapidminer</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/919/1*rBrtGo9so5B8d3BiTjDtkQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*njRy1vtOXENYheEIBz0crg.png" /></figure><p>6. Feature scaler (use StandardScaler) by Python</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1015/1*LpeRUEoAiTtk3S2BLkJo_A.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1015/1*hvANhOMvAjI35E5bmo8OOw.png" /></figure><p>6. Feature scaler (use Z-transformation) by Rapidminer</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/921/1*B4jynpTlFybVTNt7nLPnyg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*54fc_Kr4dKtkhKGlQY0Wcg.png" /></figure><p>7. Select and train model by Python</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GAYIuU5YhRljemkyzXPVnA.png" /></figure><p>7. Select and train model by Rapidminer</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/921/1*-Cdj9ZzJwZrT1Gs-ROBfLA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2utDN44LpdbrpcnvTQ89hw.png" /></figure><p>8. Perfomance by Python</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4Mu3EX_9tViohI5WGdlmDQ.png" /></figure><p>8. Performance by Rapidminer</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/604/1*riueyvgbdmjpdG3-_Vl0QQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/604/1*eA8wMfzmg4SIdZealO0osw.png" /></figure><p>สรุปก็จะประมาณนี้ครับ การใช้ Python และ Rapidminer สำหรับงาน Machine Learining ซึ่งผมต้องศึกษาเพิ่มเติมอีกมาก ไว้มีความรู้มากๆๆๆๆกว่านี้ แล้วจะมา Share กันอีกนะครับ</p><p>ขอบคุณที่อ่านจนจบนะครับ ผิดพลาดประการใด comment เลยนะครับ จะได้ปรับปรุงความรู้กันให้มากกว่านี้ครับ</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=62a423be04e6" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>