<?xml version='1.0' encoding='UTF-8'?><?xml-stylesheet href="http://www.blogger.com/styles/atom.css" type="text/css"?><feed xmlns='http://www.w3.org/2005/Atom' xmlns:openSearch='http://a9.com/-/spec/opensearchrss/1.0/' xmlns:blogger='http://schemas.google.com/blogger/2008' xmlns:georss='http://www.georss.org/georss' xmlns:gd="http://schemas.google.com/g/2005" xmlns:thr='http://purl.org/syndication/thread/1.0'><id>tag:blogger.com,1999:blog-2451586136929499042</id><updated>2024-09-07T02:07:34.283+07:00</updated><category term="writing"/><category term="activity based"/><category term="cross validation"/><category term="gis"/><category term="opensource"/><category term="Bangkok"/><category term="Endnote"/><category term="ITS"/><category term="MATLAB"/><category term="bioinformatics"/><category term="book"/><category term="bowman"/><category term="bus"/><category term="calibration"/><category term="citation"/><category term="classification tree"/><category term="clustering"/><category term="computing"/><category term="corsim"/><category term="data mining"/><category term="dynamic traffic assignment"/><category term="export"/><category term="figure"/><category term="graphic"/><category term="ieee"/><category term="integration"/><category term="k-means"/><category term="microscopic simulation"/><category term="sequence"/><category term="statistics"/><category term="talk"/><category term="traffic simulation"/><category term="word"/><title type='text'>TranSIIT</title><subtitle type='html'>Transport Research Group at SIIT</subtitle><link rel='http://schemas.google.com/g/2005#feed' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/posts/default'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default?redirect=false'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/'/><link rel='hub' href='http://pubsubhubbub.appspot.com/'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><generator version='7.00' uri='http://www.blogger.com'>Blogger</generator><openSearch:totalResults>23</openSearch:totalResults><openSearch:startIndex>1</openSearch:startIndex><openSearch:itemsPerPage>25</openSearch:itemsPerPage><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-5477923838118497774</id><published>2010-11-20T15:36:00.001+07:00</published><updated>2010-11-20T15:36:53.366+07:00</updated><category scheme="http://www.blogger.com/atom/ns#" term="Endnote"/><category scheme="http://www.blogger.com/atom/ns#" term="writing"/><title type='text'>Endnote FAQ [In-text Citation]</title><content type='html'>&lt;div class=&quot;separator&quot; style=&quot;clear: both; text-align: center;&quot;&gt;&lt;a href=&quot;http://testpilot.library.vanderbilt.edu/testpilot/images/endnote-zotero.gif&quot; imageanchor=&quot;1&quot; style=&quot;margin-left: 1em; margin-right: 1em;&quot;&gt;&lt;img border=&quot;0&quot; src=&quot;http://testpilot.library.vanderbilt.edu/testpilot/images/endnote-zotero.gif&quot; /&gt;&lt;/a&gt;&lt;/div&gt;&lt;br /&gt;
Inserting in-text citation of different styles into the same document is also one of my current problem when using bibliographic management program such as Zotero and Endnote.&lt;br /&gt;
&lt;br /&gt;
Here is some discussion from Endnote&#39;s website. Instead of using manual editing you can use&amp;nbsp;some techniques but not the perfect solution though.&lt;br /&gt;
&lt;br /&gt;
&lt;a href=&quot;http://community.thomsonreuters.com/t5/EndNote-How-To/Formatting-in-text-citation-brackets-paranthesis/m-p/7290&quot;&gt;http://community.thomsonreuters.com/t5/EndNote-How-To/Formatting-in-text-citation-brackets-paranthesis/m-p/7290&lt;/a&gt;</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/5477923838118497774/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/5477923838118497774' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/5477923838118497774'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/5477923838118497774'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2010/11/endnote-faq-in-text-citation.html' title='Endnote FAQ [In-text Citation]'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-1169901330929309744</id><published>2010-10-07T10:48:00.001+07:00</published><updated>2010-10-07T10:49:18.209+07:00</updated><category scheme="http://www.blogger.com/atom/ns#" term="writing"/><title type='text'>A Web Site to Check the English Definition and Usage</title><content type='html'>&lt;a href=&quot;http://www.wordreference.com/definition/&quot;&gt;http://www.wordreference.com/definition/&lt;/a&gt;&lt;br /&gt;
&lt;br /&gt;
When doubted about the usage of words such as&lt;br /&gt;
&lt;br /&gt;
as follows or as following&lt;br /&gt;
base on or based on&lt;br /&gt;
disability or disabled, etc.&lt;br /&gt;
&lt;br /&gt;
Just type in the words and see the result in discussion forum.&lt;br /&gt;
There are many English experts there.</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/1169901330929309744/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/1169901330929309744' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/1169901330929309744'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/1169901330929309744'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2010/10/web-site-to-check-english-definition.html' title='A Web Site to Check the English Definition and Usage'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-6244322677789260793</id><published>2010-10-06T09:17:00.000+07:00</published><updated>2010-10-06T09:17:00.621+07:00</updated><title type='text'>Table of mathematical symbols</title><content type='html'>This is useful source of information about mathematical notations or symbols reading and usage.&lt;br /&gt;
&lt;br /&gt;
&lt;a href=&quot;http://en.wikipedia.org/wiki/Table_of_mathematical_symbols&quot;&gt;http://en.wikipedia.org/wiki/Table_of_mathematical_symbols&lt;/a&gt;&lt;br /&gt;
&lt;br /&gt;
&lt;div class=&quot;separator&quot; style=&quot;clear: both; text-align: center;&quot;&gt;&lt;a href=&quot;http://upload.wikimedia.org/wikipedia/commons/e/eb/Math.svg&quot; imageanchor=&quot;1&quot; style=&quot;clear: left; float: left; margin-bottom: 1em; margin-right: 1em;&quot;&gt;&lt;img border=&quot;0&quot; height=&quot;320&quot; src=&quot;http://upload.wikimedia.org/wikipedia/commons/e/eb/Math.svg&quot; width=&quot;320&quot; /&gt;&lt;/a&gt;&lt;/div&gt;</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/6244322677789260793/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/6244322677789260793' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/6244322677789260793'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/6244322677789260793'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2010/10/table-of-mathematical-symbols.html' title='Table of mathematical symbols'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-5349990512290744439</id><published>2010-09-12T00:00:00.000+07:00</published><updated>2010-09-12T00:00:18.525+07:00</updated><category scheme="http://www.blogger.com/atom/ns#" term="activity based"/><category scheme="http://www.blogger.com/atom/ns#" term="bowman"/><title type='text'>Dr.Bowman&#39;s Web Site</title><content type='html'>A good source of information to build activity based travel demand model&lt;br /&gt;
&lt;br /&gt;
&lt;a href=&quot;http://jbowman.net/&quot;&gt;http://jbowman.net/&lt;/a&gt;&lt;br /&gt;
&lt;br /&gt;
&lt;div class=&quot;separator&quot; style=&quot;clear: both; text-align: center;&quot;&gt;&lt;a href=&quot;http://jbowman.net/image002.jpg&quot; imageanchor=&quot;1&quot; style=&quot;clear: left; float: left; margin-bottom: 1em; margin-right: 1em;&quot;&gt;&lt;img border=&quot;0&quot; src=&quot;http://jbowman.net/image002.jpg&quot; /&gt;&lt;/a&gt;&lt;/div&gt;</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/5349990512290744439/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/5349990512290744439' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/5349990512290744439'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/5349990512290744439'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2010/09/drbowmans-web-site.html' title='Dr.Bowman&#39;s Web Site'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-6723483080505049838</id><published>2010-09-10T21:04:00.000+07:00</published><updated>2010-09-10T21:04:58.053+07:00</updated><category scheme="http://www.blogger.com/atom/ns#" term="export"/><category scheme="http://www.blogger.com/atom/ns#" term="figure"/><category scheme="http://www.blogger.com/atom/ns#" term="graphic"/><category scheme="http://www.blogger.com/atom/ns#" term="MATLAB"/><category scheme="http://www.blogger.com/atom/ns#" term="word"/><title type='text'>เคล็ดลับการเซฟไฟล์ภาพจาก MATLAB ไปใช้กับโปรแกรมอื่นๆ</title><content type='html'>หากเราสร้าง Graph หรือ Figure ใน MATLAB แล้วต้องการเซฟไปใช้กับโปรแกรมอื่นๆให้ภาพออกมาสวยงามทำอย่างไร&lt;br /&gt;
&lt;br /&gt;
ถ้าต้องการให้พิมพ์ (Hard copy) ออกมาสวยงาม ให้เลือก File &amp;gt; Save As แล้วเลือกเป็น EPS ไฟล์&lt;br /&gt;
จากนั้นหากต้องการนำไปใช้ใน Word ตัวโปรแกรม Word จะมีตัวเลือกให้ Import ตัว EPS ไฟล์เข้าไปได้&lt;br /&gt;
&lt;br /&gt;
การแสดงผลบนหน้าจอจะไม่ค่อยชัดเจนมากนัก แต่ว่าเวลาพิมพ์ออกมาจะสวยสุดครับ&lt;br /&gt;
&lt;br /&gt;
ข้อมูลเกี่ยวกับ EPS file format หาอ่านได้ใน internet ทั่วไป</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/6723483080505049838/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/6723483080505049838' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/6723483080505049838'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/6723483080505049838'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2010/09/matlab.html' title='เคล็ดลับการเซฟไฟล์ภาพจาก MATLAB ไปใช้กับโปรแกรมอื่นๆ'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-6886888768074055656</id><published>2010-09-07T14:29:00.001+07:00</published><updated>2010-09-07T14:29:52.019+07:00</updated><category scheme="http://www.blogger.com/atom/ns#" term="corsim"/><category scheme="http://www.blogger.com/atom/ns#" term="traffic simulation"/><title type='text'>Traffic Analysis Tools by FHWA</title><content type='html'>Source of information for traffic analysis tool and traffic simulation.&lt;br /&gt;
&lt;a href=&quot;http://ops.fhwa.dot.gov/trafficanalysistools/&quot;&gt;http://ops.fhwa.dot.gov/trafficanalysistools/&lt;/a&gt;&lt;br /&gt;
&lt;br /&gt;
In particular, Volume IV is introduction to CORSIM.</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/6886888768074055656/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/6886888768074055656' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/6886888768074055656'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/6886888768074055656'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2010/09/traffic-analysis-tools-by-fhwa.html' title='Traffic Analysis Tools by FHWA'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-5142854119684584316</id><published>2010-09-05T08:06:00.003+07:00</published><updated>2010-09-05T08:20:03.934+07:00</updated><category scheme="http://www.blogger.com/atom/ns#" term="classification tree"/><category scheme="http://www.blogger.com/atom/ns#" term="cross validation"/><title type='text'>ปัญหา (เล็กน้อยๆ) ที่อาจเกิดขึ้นจากการทำ Cross Validation กับข้อมูลที่มี Categorical Variables</title><content type='html'>ถ้าทำ 10-fold Cross Validation มันอาจบังเิอิญได้ว่าตอนที่เราเลือก 90% data มาทำโมเดล แล้วไปทดสอบกับ 10% ที่เหลือนั้น&lt;br /&gt;
&lt;br /&gt;
เจ้า 90% data ที่นำมาสร้างโมเดล หากตัว categorical variable ที่เรานำมาสร้างโมเดลนั้น มันไม่ครอบคลุมทั้งหมด&lt;br /&gt;
&lt;br /&gt;
เช่น ถ้าตัวแปรเป็นสถานะภาพในครอบครัว เช่น ผู้อาศัย แต่ตอนทำโมเดล 90% ไม่มีข้อมูลที่เป็นผู้อาศัยเลย ดังนั้นเวลาเรานำไปทดสอบกับข้อมูล 10% ที่อาจมีข้อมูลที่มีสถานภาพเป็นผู้อาศัย มันก็อาจเกิดข้อผิดพลาดได้ (สำหรับ Classification Tree ใน MATLAB จะขึ้น Error แต่ถ้าเป็นโปรแกรมที่เขียนขึ้นเองอาจไม่ขึ้น error เพราะตัวโค๊ดอาจจะปัดให้เป็น Branch Not in Category เลยแต่ผลที่ได้อาจไม่น่าเชื่อถือ ถ้าจริงๆแล้วมันควรตกอีกด้านหนึ่ง แต่เราไม่ได้มี data ไปสร้างโมเดลให้มัน)&lt;br /&gt;
&lt;br /&gt;
วิธีแก้ไขตอนนี้ยังคิดไม่ออกทั้งหมด แต่ว่าขึ้นกับการสุ่มตัว sample ถ้าสุ่มได้ครอบคลุมก็จะลดปัญหาไปได้ (แต่ในกรณีข้อมูลมีตัวแปร category เยอะ และมีค่าหลากหลาย อาจสุ่มได้ยากกว่า) แต่ถ้ามานั่งเลือกเอง data เยอะๆ คงจะเสียเวลาและ งงๆ อาจต้องอ่านงานวิจัยเพิ่มวิธีการสุ่มให้ครอบคลุม&lt;br /&gt;
&lt;br /&gt;
อีกวิธีหนึ่งที่อาจลดปัญหาได้บ้าง คือการใช้ 5-fold cross classification แทน (80%/20%) มีความน่าจะเป็นที่ปัญหาจะลดลงได้ แต่ก็ไม่รับประกัน (ใน MATLAB ปกติจะเป็น 10-fold)&lt;br /&gt;
&lt;br /&gt;
อีกวิธีที่คิดได้คือเลือกข้อมูลที่เป็น category ย่อย ที่มีจำนวน records น้อย ออกมาก่อนแล้ว assign ให้มันอยู่ในกลุ่ม 90% เสมอ ก็จะไม่เกิดกรณีเจอ unrecognized categorical variable ตอนทดสอบ (แต่ถ้าใช้ MATLAB ต้องทำแบบ manual เพราะโค๊ดมันคงไม่ทำให้ หรือไม่ก็ต้องไปแ้ก้ตัวโค๊ด)&lt;br /&gt;
&lt;br /&gt;
วิธีสุดท้าย อาจไม่ใช้ v-fold cross validation เลย ถ้าคิดว่าข้อมูลเยอะพอ คือเลือกข้อมูลที่ครบถ้วน แล้วทดสอบทีเดียว&lt;br /&gt;
&lt;br /&gt;
สุดท้ายนี้ปัญหานี้จะไม่เกิดขึ้นถ้าเป็น continuous variable เพราะโดยปกติจะแบ่ง Branch ซ้าย-ขวา โดยใช้เครื่องหมายเช่น น้อยกว่าหรือเท่ากับ และมากกว่า ซึ่งจะครอบคลุม Range ของข้อมูลทั้งหมดอยู่แล้ว แต่ก็ยังไม่ป้องกันปัญหาที่อาจจะเกิดจากการไม่ include หรือ neglect ตัวแปรบางตัวที่ไม่ได้มีอยู่ใน 90% data อยู่ดี&lt;br /&gt;
&lt;br /&gt;
คิมครับ ลองเช็คดูว่าโค๊ด Python ของคิม ถ้าเจอกรณีนี้มันจะขึ้น Error หรือไม่อย่างไรครับ</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/5142854119684584316/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/5142854119684584316' title='2 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/5142854119684584316'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/5142854119684584316'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2010/09/cross-validation-categorical-variables.html' title='ปัญหา (เล็กน้อยๆ) ที่อาจเกิดขึ้นจากการทำ Cross Validation กับข้อมูลที่มี Categorical Variables'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><thr:total>2</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-542398688535961427</id><published>2010-08-31T16:18:00.006+07:00</published><updated>2010-08-31T23:04:56.095+07:00</updated><category scheme="http://www.blogger.com/atom/ns#" term="cross validation"/><category scheme="http://www.blogger.com/atom/ns#" term="data mining"/><title type='text'>วิธีง่ายๆในการแบ่ง Data ออกเป็น k ส่วน</title><content type='html'>สมมุติว่าเรามีข้อมูลเป็น record อยู่ทั้งหมดจำนวน n records&lt;br /&gt;
แล้วเราต้องการแบ่งข้อมูลนี้ออกเป็น k ส่วน&lt;br /&gt;
&lt;br /&gt;
ตัวอย่างเช่น ถ้าเราต้องการทำ V-fold Cross-Validation เช่นต้องการแบ่งข้อมูลจำนวน 2000 ระเบียนออกเป็น 10 ส่วนโดยส่วนหนึ่งเอาไว้สร้าง model กับ อีกส่วนไว้ validate model เป็นต้น&lt;br /&gt;
&lt;br /&gt;
ในกรณีนี้ n = 2000 และ k = 10 เป็นต้น&lt;br /&gt;
&lt;br /&gt;
มีวิธีทำง่ายๆ ใน MATLAB ดังนี้&lt;br /&gt;
&lt;ol&gt;&lt;li&gt;สร้าง random number ขึ้นมา 2000 ตัวก่อน, &amp;nbsp; rand(2000,1)&lt;/li&gt;
&lt;li&gt;scale up มันด้วย k, rand(2000,1)*k&lt;/li&gt;
&lt;li&gt;ปัดตัวเลขขึ้นเป็น integer ถัดไป, ceil(rand(2000,1)*k)&lt;/li&gt;
&lt;/ol&gt;&lt;div&gt;สรุปคำสั่งที่ต้องใช้คือ&lt;/div&gt;&lt;div&gt;&lt;br /&gt;
&lt;/div&gt;&lt;div&gt;ceil(rand(n,1)*k)&lt;/div&gt;&lt;div&gt;&lt;br /&gt;
&lt;/div&gt;&lt;div&gt;ตัวอย่างการแบ่งข้อมูล 10 ระเบียนออกเป็น 3 กลุ่ม&lt;br /&gt;
&lt;br /&gt;
&amp;gt;&amp;gt; ceil(rand(10,1)*3)&lt;br /&gt;
&lt;br /&gt;
ans =&lt;br /&gt;
&lt;br /&gt;
&amp;nbsp;&amp;nbsp; &amp;nbsp; 2&lt;br /&gt;
&amp;nbsp;&amp;nbsp; &amp;nbsp; 1&lt;br /&gt;
&amp;nbsp;&amp;nbsp; &amp;nbsp; 1&lt;br /&gt;
&amp;nbsp;&amp;nbsp; &amp;nbsp; 1&lt;br /&gt;
&amp;nbsp;&amp;nbsp; &amp;nbsp; 3&lt;br /&gt;
&amp;nbsp;&amp;nbsp; &amp;nbsp; 1&lt;br /&gt;
&amp;nbsp;&amp;nbsp; &amp;nbsp; 3&lt;br /&gt;
&amp;nbsp;&amp;nbsp; &amp;nbsp; 2&lt;br /&gt;
&amp;nbsp;&amp;nbsp; &amp;nbsp; 3&lt;br /&gt;
&amp;nbsp;&amp;nbsp; &amp;nbsp; 2&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
Note: แต่คิดไปคิดมา ถ้าไม่ต้องการ Random ละก็ ให้มัน mod ตัวเลข 1 ถึง n ก็จะได้เหมือนกันครับ&lt;br /&gt;
เช่น mod(1:10,3)+1 จะไ้ด้ผลลัพธ์ประมาณนี้ครับ&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;nbsp;2 &amp;nbsp; &amp;nbsp; 3 &amp;nbsp; &amp;nbsp; 1 &amp;nbsp; &amp;nbsp; 2 &amp;nbsp; &amp;nbsp; 3 &amp;nbsp; &amp;nbsp; 1 &amp;nbsp; &amp;nbsp; 2 &amp;nbsp; &amp;nbsp; 3 &amp;nbsp; &amp;nbsp; 1 &amp;nbsp; &amp;nbsp; 2&lt;br /&gt;
&lt;div&gt;&lt;br /&gt;
&lt;/div&gt;&lt;/div&gt;</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/542398688535961427/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/542398688535961427' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/542398688535961427'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/542398688535961427'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2010/08/data-k.html' title='วิธีง่ายๆในการแบ่ง Data ออกเป็น k ส่วน'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-7542572645754968084</id><published>2010-06-02T16:15:00.000+07:00</published><updated>2010-06-02T16:15:45.242+07:00</updated><title type='text'>Some Tutorial for Creating Scoring Matrices</title><content type='html'>&lt;a href=&quot;http://www.biorecipes.com/Dayhoff/code.html&quot;&gt;http://www.biorecipes.com/Dayhoff/code.html&lt;/a&gt;&amp;nbsp;(for PAM)&lt;br /&gt;
&lt;a href=&quot;http://www.it.cabri.org/training/bcd/Curric/PrwAli/nodeD.html&quot;&gt;http://www.it.cabri.org/training/bcd/Curric/PrwAli/nodeD.html&lt;/a&gt;&amp;nbsp;(for PAM and BLOSUM, etc)</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/7542572645754968084/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/7542572645754968084' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/7542572645754968084'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/7542572645754968084'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2010/06/some-tutorial-for-creating-scoring.html' title='Some Tutorial for Creating Scoring Matrices'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-7627762284272849134</id><published>2010-05-21T15:26:00.000+07:00</published><updated>2010-05-21T15:26:15.578+07:00</updated><category scheme="http://www.blogger.com/atom/ns#" term="bioinformatics"/><category scheme="http://www.blogger.com/atom/ns#" term="book"/><category scheme="http://www.blogger.com/atom/ns#" term="sequence"/><category scheme="http://www.blogger.com/atom/ns#" term="statistics"/><title type='text'>Recommended Book: Statistical methods in bioinformatics</title><content type='html'>This book is recommended for studying about sequence, statistical methods in bioinformatics.&lt;br /&gt;
&lt;br /&gt;
&lt;span class=&quot;Apple-style-span&quot; style=&quot;font-family: Arial, sans-serif; font-size: 18px; font-weight: bold; white-space: nowrap;&quot;&gt;Statistical methods in bioinformatics&lt;/span&gt;&lt;br /&gt;
&lt;span class=&quot;Apple-style-span&quot; style=&quot;font-family: Arial, sans-serif; font-size: 18px; font-weight: bold; white-space: nowrap;&quot;&gt;&lt;span class=&quot;Apple-style-span&quot; style=&quot;font-weight: normal;&quot;&gt;by&amp;nbsp;&lt;/span&gt;&lt;span class=&quot;Apple-style-span&quot; style=&quot;border-collapse: collapse; font-family: arial, helvetica, clean, sans-serif; font-size: 13px; line-height: 16px; white-space: normal;&quot;&gt;Warren J. Ewens Gregory Grant&lt;/span&gt;&lt;/span&gt;&lt;br /&gt;
&lt;span class=&quot;Apple-style-span&quot; style=&quot;font-family: Arial, sans-serif; font-size: 18px; font-weight: bold; white-space: nowrap;&quot;&gt;&lt;span class=&quot;Apple-style-span&quot; style=&quot;border-collapse: collapse; font-family: arial, helvetica, clean, sans-serif; font-size: 13px; line-height: 16px; white-space: normal;&quot;&gt;&lt;br /&gt;
&lt;/span&gt;&lt;/span&gt;&lt;br /&gt;
&lt;div class=&quot;separator&quot; style=&quot;clear: both; text-align: center;&quot;&gt;&lt;a href=&quot;https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhC3anK-MpHxo5Yjdz6VV6t_prFs4dtemy8kG9f1h8m6neRzBCHZ_9HUaiWFUZEk10Cp12GUcqd7TZ1B6nik1YK-jlTOcqMblVOFTjLkI2dv1WMZ-2QwIs48MVpMekSDxZJMPIINPFyb6uC/s1600/0387952292.01.LZZZZZZZ.jpg&quot; imageanchor=&quot;1&quot; style=&quot;clear: left; float: left; margin-bottom: 1em; margin-right: 1em;&quot;&gt;&lt;img border=&quot;0&quot; src=&quot;https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhC3anK-MpHxo5Yjdz6VV6t_prFs4dtemy8kG9f1h8m6neRzBCHZ_9HUaiWFUZEk10Cp12GUcqd7TZ1B6nik1YK-jlTOcqMblVOFTjLkI2dv1WMZ-2QwIs48MVpMekSDxZJMPIINPFyb6uC/s320/0387952292.01.LZZZZZZZ.jpg&quot; /&gt;&lt;/a&gt;&lt;/div&gt;&lt;span class=&quot;Apple-style-span&quot; style=&quot;font-family: Arial, sans-serif; font-size: 18px; font-weight: bold; white-space: nowrap;&quot;&gt;&lt;span class=&quot;Apple-style-span&quot; style=&quot;border-collapse: collapse; font-family: arial, helvetica, clean, sans-serif; font-size: 13px; line-height: 16px; white-space: normal;&quot;&gt;&lt;br /&gt;
&lt;/span&gt;&lt;/span&gt;</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/7627762284272849134/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/7627762284272849134' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/7627762284272849134'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/7627762284272849134'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2010/05/recommended-book-statistical-methods-in.html' title='Recommended Book: Statistical methods in bioinformatics'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><media:thumbnail xmlns:media="http://search.yahoo.com/mrss/" url="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhC3anK-MpHxo5Yjdz6VV6t_prFs4dtemy8kG9f1h8m6neRzBCHZ_9HUaiWFUZEk10Cp12GUcqd7TZ1B6nik1YK-jlTOcqMblVOFTjLkI2dv1WMZ-2QwIs48MVpMekSDxZJMPIINPFyb6uC/s72-c/0387952292.01.LZZZZZZZ.jpg" height="72" width="72"/><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-8397229323992808977</id><published>2010-05-20T15:24:00.000+07:00</published><updated>2010-05-20T15:24:05.887+07:00</updated><category scheme="http://www.blogger.com/atom/ns#" term="clustering"/><category scheme="http://www.blogger.com/atom/ns#" term="k-means"/><title type='text'>หลักการเลือกจำนวน clusters ที่เหมาะสมสำหรับการทำ Clustering</title><content type='html'>ในการแบ่งกลุ่ม clusters นั้นบางครั้งเราอาจจะต้องกำหนดจำนวน clusters ที่เหมาะสม โดยเป็นค่า parameter หนึ่งของการคำนวณ&lt;br /&gt;
เช่นการทำ k-means clustering ต้องกำหนดจำนวนกลุ่มให้ algorithm ก่อน&lt;br /&gt;
ดูหลักการได้ใน&lt;br /&gt;
&lt;a href=&quot;http://en.wikipedia.org/wiki/Determining_the_number_of_clusters_in_a_data_set&quot;&gt;หลักการเลือกจำนวน clusters ที่เหมาะสมสำหรับการทำ clustering (wikipedia)&lt;/a&gt;&lt;br /&gt;
&lt;br /&gt;
Note: ถ้าเอาไป apply กับ CART ก็อาจใช้กำหนดจำนวน node หรือ ระดับการ prune ที่เหมาะสมได้&lt;br /&gt;
&lt;br /&gt;
&lt;div class=&quot;separator&quot; style=&quot;clear: both; text-align: center;&quot;&gt;&lt;a href=&quot;https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgpRlJPqu-UmsznHJjJSWIfuswttqrFno0L88U7eJHSgK1G8kpypBO44910VdNIop5WplepAx3hn9g0y9DSL1HFD2stwH4fERQ5PNJr_e5RwyGlquwBK2dBime1oCcjwK-My1ij_BQeKUga/s1600/clusters.gif&quot; imageanchor=&quot;1&quot; style=&quot;clear: left; float: left; margin-bottom: 1em; margin-right: 1em;&quot;&gt;&lt;img border=&quot;0&quot; src=&quot;https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgpRlJPqu-UmsznHJjJSWIfuswttqrFno0L88U7eJHSgK1G8kpypBO44910VdNIop5WplepAx3hn9g0y9DSL1HFD2stwH4fERQ5PNJr_e5RwyGlquwBK2dBime1oCcjwK-My1ij_BQeKUga/s320/clusters.gif&quot; /&gt;&lt;/a&gt;&lt;/div&gt;</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/8397229323992808977/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/8397229323992808977' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/8397229323992808977'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/8397229323992808977'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2010/05/clusters-clustering.html' title='หลักการเลือกจำนวน clusters ที่เหมาะสมสำหรับการทำ Clustering'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><media:thumbnail xmlns:media="http://search.yahoo.com/mrss/" url="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgpRlJPqu-UmsznHJjJSWIfuswttqrFno0L88U7eJHSgK1G8kpypBO44910VdNIop5WplepAx3hn9g0y9DSL1HFD2stwH4fERQ5PNJr_e5RwyGlquwBK2dBime1oCcjwK-My1ij_BQeKUga/s72-c/clusters.gif" height="72" width="72"/><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-6694657121526198558</id><published>2008-11-03T13:46:00.002+07:00</published><updated>2008-11-03T13:51:04.885+07:00</updated><category scheme="http://www.blogger.com/atom/ns#" term="citation"/><category scheme="http://www.blogger.com/atom/ns#" term="ieee"/><category scheme="http://www.blogger.com/atom/ns#" term="writing"/><title type='text'>Citation IEEE Style (Explained in Thai)</title><content type='html'>&lt;a href=&quot;http://www.doc.eng.cmu.ac.th/%7Etrasapong/Cit.html&quot;&gt;Citation IEEE Style (Explained in Thai)&lt;/a&gt;&lt;a onblur=&quot;try {parent.deselectBloggerImageGracefully();} catch(e) {}&quot; href=&quot;https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEguIF7fAteYm6zibYmOkZCDfGCseZEz3QB5j9Ef3AS8RWu9swx4dyD_Z0cZvl9ErqE9jhjq1mViJ_t6Hdkt5tfYxkibLwXhyNlYSNcl0jWb97Zr1gyXRU3tg6YthB6TrzPAy1j37h2Bs2m2/s1600-h/IEEE+large.jpg&quot;&gt;&lt;img style=&quot;margin: 0pt 10px 10px 0pt; float: left; cursor: pointer; width: 200px; height: 71px;&quot; src=&quot;https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEguIF7fAteYm6zibYmOkZCDfGCseZEz3QB5j9Ef3AS8RWu9swx4dyD_Z0cZvl9ErqE9jhjq1mViJ_t6Hdkt5tfYxkibLwXhyNlYSNcl0jWb97Zr1gyXRU3tg6YthB6TrzPAy1j37h2Bs2m2/s200/IEEE+large.jpg&quot; alt=&quot;&quot; id=&quot;BLOGGER_PHOTO_ID_5264319918304571090&quot; border=&quot;0&quot; /&gt;&lt;/a&gt;</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/6694657121526198558/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/6694657121526198558' title='1 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/6694657121526198558'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/6694657121526198558'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2008/11/citation-ieee-style-explained-in-thai.html' title='Citation IEEE Style (Explained in Thai)'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><media:thumbnail xmlns:media="http://search.yahoo.com/mrss/" url="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEguIF7fAteYm6zibYmOkZCDfGCseZEz3QB5j9Ef3AS8RWu9swx4dyD_Z0cZvl9ErqE9jhjq1mViJ_t6Hdkt5tfYxkibLwXhyNlYSNcl0jWb97Zr1gyXRU3tg6YthB6TrzPAy1j37h2Bs2m2/s72-c/IEEE+large.jpg" height="72" width="72"/><thr:total>1</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-1611446135949449376</id><published>2008-11-02T14:18:00.004+07:00</published><updated>2008-11-02T14:34:45.003+07:00</updated><category scheme="http://www.blogger.com/atom/ns#" term="writing"/><title type='text'>Table of mathematical symbols</title><content type='html'>We better study the way to express things mathematically correct.&lt;br /&gt;&lt;br /&gt;For reference:&lt;br /&gt;&lt;a href=&quot;http://en.wikipedia.org/wiki/Table_of_mathematical_symbols&quot;&gt;http://en.wikipedia.org/wiki/Table_of_mathematical_symbols&lt;/a&gt;&lt;br /&gt;&lt;br /&gt;&lt;a onblur=&quot;try {parent.deselectBloggerImageGracefully();} catch(e) {}&quot; href=&quot;http://www.fotosearch.com/bthumb/phd/PHD420/BS32043.jpg&quot;&gt;&lt;img style=&quot;margin: 0pt 10px 10px 0pt; float: left; cursor: pointer; width: 124px; height: 170px;&quot; src=&quot;http://www.fotosearch.com/bthumb/phd/PHD420/BS32043.jpg&quot; alt=&quot;&quot; border=&quot;0&quot; /&gt;&lt;/a&gt;</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/1611446135949449376/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/1611446135949449376' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/1611446135949449376'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/1611446135949449376'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2008/11/table-of-mathematical-symbols.html' title='Table of mathematical symbols'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-9138498703404957332</id><published>2008-07-31T21:34:00.001+07:00</published><updated>2008-07-31T21:37:52.282+07:00</updated><title type='text'>A Guidebook for Measuring Crime in Public Housing with Geographic Information Systems (August 1999, 33 p.)</title><content type='html'>บังเอิญก่อนกลับบ้านลอง Google เล่นเกี่ยวกับหัวข้อ Crime Mapping ที่วินทำอยู่ครับ แล้วก็เป็นหัวข้อที่ผมสนใจอยู่เหมือนกัน ก็ไปเจอ Guidebook ตัวนี้ครับมีรายละเอียดการวิเคราะห์&lt;br /&gt;ที่ผมว่าน่าจะนำไปเป็นแนวทางในการทำ research ของวินในเบื้องต้นได้ครับ เป็น word doc&lt;br /&gt;ทั้งหมด 30 กว่าหน้าครับน่าสนใจดี ลิงค์ตามนี้เลยครับ&lt;br /&gt;&lt;br /&gt;&lt;a href=&quot;http://www.huduser.org/publications/pubasst/gis.html&quot;&gt;A Guidebook for Measuring Crime in Public Housing with Geographic Information Systems &lt;i&gt;(August 1999, 33 p.)&lt;/i&gt;&lt;/a&gt;</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/9138498703404957332/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/9138498703404957332' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/9138498703404957332'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/9138498703404957332'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2008/07/guidebook-for-measuring-crime-in-public.html' title='A Guidebook for Measuring Crime in Public Housing with Geographic Information Systems (August 1999, 33 p.)'/><author><name>คิม</name><uri>http://www.blogger.com/profile/09715683525677301971</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='31' height='22' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiVqiQH38LYmJPmJm4ByuUPBTd-azQRs5CBS6wJOLsasghtB5GgBSfVdMUfdlULQzJKmB8bi-FIIQju3qji3w03sypZX7FViKw1CLbR85wMUY76AExFnpTyS5dIDXrotw/s220/551000003897601.jpg'/></author><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-7674494919574761677</id><published>2008-07-24T15:33:00.002+07:00</published><updated>2008-07-24T15:35:21.271+07:00</updated><title type='text'>Travel time estimation on a freeway using Discrete Time Markov Chains</title><content type='html'>&lt;div xmlns=&quot;http://www.w3.org/1999/xhtml&quot;&gt;จำไม่ได้ว่าเคยคุยกับคิมหรือเปล่า มีคนเอาไปทำแล้ว&lt;br /&gt;&lt;br /&gt;ถ้าจำไม่ผิดเราเคยมีไอเดียคร่าวๆว่าน่าจะใช้ markov chain ทำนายสภาพจราจรบนเน็ตเวิร์ค ณ เวลาต่างๆกันได้&lt;br /&gt;&lt;br /&gt;&lt;a href=&quot;http://www.sciencedirect.com/science?_ob=ArticleURL&amp;amp;_udi=B6V99-4PYRK6W-1&amp;amp;_user=578664&amp;amp;_rdoc=1&amp;amp;_fmt=&amp;amp;_orig=search&amp;amp;_sort=d&amp;amp;view=c&amp;amp;_acct=C000029478&amp;amp;_version=1&amp;amp;_urlVersion=0&amp;amp;_userid=578664&amp;amp;md5=c63a87a06627dbd23b5a7d6ab8174a9e&quot;&gt;paper&lt;/a&gt; นี้ เขาใช้ discrete time markov chain หา estimated route&#39;s travel time บนทางด่วนได้&lt;br /&gt;&lt;br /&gt;ถ้าลองอ่านดูจะพบว่าไอเดียดีครับ ทำไม่ยาก น่าจะทำเพิ่มเติมได้อีก&lt;br /&gt;&lt;br /&gt;&lt;img style=&quot;width: 605px; height: 181px;&quot; title=&quot;Image&quot; alt=&quot;Image&quot; src=&quot;http://www.sciencedirect.com/cache/MiamiImageURL/B6V99-4PYRK6W-1-1/0?wchp=dGLbVtz-zSkWb&quot; align=&quot;bottom&quot; border=&quot;0&quot; /&gt;&lt;br /&gt;&lt;br /&gt;&lt;span style=&quot;font-family:sans-serif;&quot;&gt;Note: ไอเดียต่างกับเรานิดหน่อยที่เราต้องการทำนายทั้งเน็ตเวิร์ค ไม่ใช่แค่ corridor&lt;br /&gt;&lt;br /&gt;ตรงนี้ของเขาแบ่งทางด่วนช่วงหนึ่งออกเป็น 6 links&lt;br /&gt;แต่ละ link ให้มี state ได้คือ 0 กับ 1 คือ รถว่าง (non-congested) กับ รถติด (congested)&lt;br /&gt;ดังนั้น สภาพของถนนทั้งหมดที่เป็นไปได้ ณ ขณะใดๆจะมีได้ 2^6 = ุ64 states&lt;br /&gt;&lt;br /&gt;แต่ในความเป็นจริงเค้าเก็บข้อมูล 4 เดือน แล้วแบ่ง transition matrix ตามช่วงเวลาอีก คือความจริง เช่น ช่วง peak เย็น อาจมีสภาพจริง ไม่ vary ถึง 64 states แต่จะสลับไปมา 20 กว่า states จากนั้นก็เก็บข้อมูลจริง แล้วหา (one-step) transition matrix  จากนั้นก็เอา matrix นั้นมาหาค่าเฉลี่ยของเวลาในการเดินทางได้ครับ โดยเขาให้ 1 step = 1 min นอกจากนี้ยังเอา ข้อมูล flow rate มาคิด โดยใช้ flow-travel time curve โดยใส่ค่า flow หา travel time (fit curve) แล้วโยนเข้าเป็น travel time โดยแบ่งเป็น congested กับ non-congested&lt;br /&gt;&lt;br /&gt;โดยรวมแล้ว idea ดีมากครับ แต่ความเห็นส่วนตัวว่าframeworkมันยังไม่ unified เท่าไหร่ เหมือนตัดๆต่อๆ เช่น eq.(6) ถ้ามีวิธีทำให้มันดีกว่านี้ก็จะดี&lt;br /&gt;&lt;br /&gt;แล้วก็ travel time มันอาจจะยังไม่ dynamics นะ เพราะว่าเอา travel time ณ ตรงเวลานั้นมาคิดเลย แต่จริงๆ กว่าจะไปถึง link ปลายๆ เวลามันก็เปลียนไปแล้ว อันนี้เหมือนเอา travel time ณ เวลาใดเวลาหนึงมาต่อๆกัน (อาจไม่ใช่ก็ได้ ต้องอ่านละเอียดอีกที) แต่ยังไงก็ตาม เจ้าของผลงานก็ claim ว่า &lt;b&gt;ค่าเฉลี่ย&lt;/b&gt; ของ travel time ที่ได้มีความใกล้เคียงกับสภาพจริง เมื่อทดสอบทางสถิติแล้ว (Note: ค่าเฉลี่ยมันก็คือภาพรวมๆ ความจริงมันน่าจะมีตัววัดอื่นๆด้วย) และสังเกตว่า ระยะทาง 7-8 ไมล์ วิ่งใช้เวลาประมาณ 7-8 นาที (ประมาณ 1 mile/นาที = 60 ไมล์/ชม) สภาพรถน่าจะไม่ติดมาก แล้วมันchallenge ยังไง???? ถ้าลองมาทำกับ urban network น่าจะท้าทายดีแต่คงมีปัญหาที่ต้องขบคิดมากกว่าแน่ๆ ลงท้ายเค้าบอกว่ามันยังมีโอกาสคิดเพิ่มเติมพวก ก่อสร้างทาง (work zone) หรือ พวก incidents ต่างๆได้&lt;br /&gt;&lt;br /&gt;กระทู้นี้โม้ยาวเหมือนกันแฮะ จบดีกว่า&lt;/span&gt;&lt;/div&gt;</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/7674494919574761677/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/7674494919574761677' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/7674494919574761677'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/7674494919574761677'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2008/07/travel-time-estimation-on-freeway-using_24.html' title='Travel time estimation on a freeway using Discrete Time Markov Chains'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-4436834015605251821</id><published>2008-07-23T12:36:00.004+07:00</published><updated>2008-07-23T12:39:43.277+07:00</updated><category scheme="http://www.blogger.com/atom/ns#" term="computing"/><category scheme="http://www.blogger.com/atom/ns#" term="opensource"/><title type='text'>GPU: Blobal Processing Unit</title><content type='html'>&lt;div xmlns=&quot;http://www.w3.org/1999/xhtml&quot;&gt;&lt;a href=&quot;http://gpu.sourceforge.net/&quot;&gt;&lt;img alt=&quot;GPU : A Gnutella Processing Unit&quot; src=&quot;http://gpu.sourceforge.net/pictures/bandeau-gpu.jpg&quot; width=&quot;800&quot; border=&quot;0&quot; height=&quot;121&quot; /&gt;&lt;/a&gt;&lt;br /&gt;&lt;br /&gt;&lt;img alt=&quot;English&quot; name=&quot;anglais&quot; src=&quot;http://gpu.sourceforge.net/pictures/Angleterre.gif&quot; width=&quot;29&quot; height=&quot;17&quot; /&gt;  GPU is a Gnutella client that allows users to share CPU-resources. GPU allows the creation of &lt;a href=&quot;http://gpu.sourceforge.net/pictures/netmapper-photo.jpg&quot;&gt;computer alliances&lt;/a&gt;.&lt;br /&gt;&lt;br /&gt;The CPU-time sharing system does not recognize privileges between users. Each person agrees to provide network resources as needed and in return is able to get CPU-cycles from other clients on the network system. Plugins extend the capabilities of client nodes and are compiled so as to take advantage of the power of the latest processors.&lt;br /&gt;&lt;br /&gt;Right now, this client allows rendering of &lt;a href=&quot;http://sourceforge.net/project/showfiles.php?group_id=58134&amp;amp;package_id=134282&amp;amp;release_id=279438&quot;&gt;Terragen movies&lt;/a&gt;. An experimental &lt;a href=&quot;http://gpu.sourceforge.net/search_engine.php&quot;&gt;search engine &lt;/a&gt;is included, too.&lt;br /&gt;&lt;br /&gt;GPU cluster size typically ranges from 5 to 15 computers. It is possible to organize ourself in teams and view statistics. Help us with your pioneer attitude in tweaking and improving this 15GHz monster :) (which is under GPL and targets only peaceful goals btw).&lt;br /&gt;&lt;br /&gt;Source: &lt;a href=&quot;http://gpu.sourceforge.net/&quot;&gt;http://gpu.sourceforge.net/&lt;/a&gt;&lt;br /&gt;&lt;br /&gt;ขอบคุณคิมครับที่ช่วย forward ข้อมูลมาให้ สนับสนุนว่าเราน่าจะ investigate ต่อครับ&lt;br /&gt;&lt;/div&gt;</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/4436834015605251821/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/4436834015605251821' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/4436834015605251821'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/4436834015605251821'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2008/07/gpu-blobal-processing-unit.html' title='GPU: Blobal Processing Unit'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-2438605520648580243</id><published>2008-07-23T12:32:00.001+07:00</published><updated>2008-07-23T12:32:12.848+07:00</updated><category scheme="http://www.blogger.com/atom/ns#" term="Bangkok"/><category scheme="http://www.blogger.com/atom/ns#" term="bus"/><category scheme="http://www.blogger.com/atom/ns#" term="gis"/><category scheme="http://www.blogger.com/atom/ns#" term="ITS"/><title type='text'>Intelligentbus.net for Bangkok</title><content type='html'>&lt;div xmlns=&#39;http://www.w3.org/1999/xhtml&#39;&gt;&lt;img height=&#39;59&#39; width=&#39;668&#39; src=&#39;http://www.intelligentbus.net/pic/title.jpg&#39;/&gt;&lt;br/&gt;&lt;br/&gt;&lt;a href=&#39;http://www.intelligentbus.net/&#39;&gt;Intelligentbus.net&lt;/a&gt; for Bangkok&lt;br/&gt;&lt;br/&gt;ต้องติดตามครับ&lt;br/&gt;ถ้าใครมีข้อมูลเพิ่มเติมช่วยใส่ลงในนี้ด้วยครับ&lt;br/&gt;&lt;/div&gt;</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/2438605520648580243/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/2438605520648580243' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/2438605520648580243'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/2438605520648580243'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2008/07/intelligentbusnet-for-bangkok.html' title='Intelligentbus.net for Bangkok'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-359163709543769689</id><published>2008-07-12T00:23:00.008+07:00</published><updated>2008-07-12T00:44:07.874+07:00</updated><category scheme="http://www.blogger.com/atom/ns#" term="gis"/><category scheme="http://www.blogger.com/atom/ns#" term="opensource"/><title type='text'>An Introduction to Open Source Geospatial Tools</title><content type='html'>&lt;span id=&quot;intelliTxt&quot;&gt; พอดีคุยๆกันเรื่อง GIS แล้วผ่านๆตาเจอ review อันนี้ครับ แต่ว่าน่าจะเก่าพอสมควรคงต้อง update ข้อมูลพอสมควรแต่ก็ดีพอใช้ได้ หาก assume ว่าโปรแกรมพวกนี้คงมีการพัฒนาอย่างต่อเนื่อง&lt;br /&gt;&lt;br /&gt;by &lt;a href=&quot;http://www.oreillynet.com/pub/au/1898&quot;&gt;Tyler Mitchell&lt;/a&gt;, author of &lt;a href=&quot;http://www.oreilly.com/catalog/webmapping/&quot;&gt;Web Mapping Illustrated&lt;/a&gt;&lt;br /&gt;06/10/2005    &lt;!-- Digital Media MPU Ad --&gt;  &lt;!-- Network MPU Ad --&gt;  &lt;!-- End ad --&gt;   &lt;!-- me --&gt;  &lt;p&gt;Geospatial technologies come in many forms, from mapping to data analysis applications. Using them has traditionally required professionals trained in geospatial information management.  While geospatial professionals have been the main users and developers of geospatial applications, the landscape has changed dramatically over the past few years.&lt;/p&gt;  &lt;p&gt;The development of open source geospatial software is an exciting part of the new geospatial landscape. Open source project offerings cover the spectrum of tools: command-line data conversion, spatially aware enterprise databases, internet mapping applications, desktop Geographic Information System (GIS) applications, geoprocessing libraries, and more. Eager developers, companies and organizations are collaborating on the new generation of geospatial technologies, providing desktop and server-side applications, APIs, and development platforms that are changing the way we work and do business. &lt;/p&gt;  &lt;p&gt;A quick glance at websites like &lt;a href=&quot;http://opensourcegis.org/&quot;&gt;OpenSourceGIS.org&lt;/a&gt; garners a massive list of tools that are freely available to you for a wide array of tasks. How do you get started?  This article introduces a handful of the applications that are available, grouped according to task.  These concepts and applications are discussed further in my book &lt;a href=&quot;http://www.oreilly.com/catalog/webmapping&quot;&gt;Web Mapping Illustrated&lt;/a&gt;.&lt;/p&gt; &lt;a href=&quot;http://conferences.oreillynet.com/where/&quot;&gt;&lt;/a&gt;    &lt;h3&gt;Data&lt;/h3&gt;  &lt;p&gt;When talking about geospatial technologies, we usually mean tools that interact with geospatial data. Interaction occurs on several fronts: creating, converting, manipulating, or visualizing data. These are the common categories of tools. Parallel to these are developer-focused tools in the form of programming libraries, allowing a deeper level of interaction.&lt;/p&gt;  &lt;p&gt;Just as wood is needed for the carpenter to use his tools, data is the raw material underlying geospatial tools. The increasing availability of geospatial information is transforming business and management in many parts of the world. Companies, communities, and individuals are realizing that they must have geospatial information to understand their world.&lt;/p&gt;  &lt;p&gt;How you store data is a critical consideration. ESRI shapefiles are in a file-based format, but for more advanced vector data management capabilities, the strongest option is the spatial extension to the PostgreSQL database, PostGIS (more on PostGIS under &lt;i&gt;Manipulation&lt;/i&gt;, below).&lt;/p&gt;  &lt;h3&gt;Creation&lt;/h3&gt;    &lt;p&gt;When data doesn&#39;t already exist, it must be created.  Some of the most traditional methods of data creation use a desktop mapping program to draw shapes on pre-existing base maps. For example, using an aerial photograph as a base, you can click on points of interest, or draw a line around a certain area. That data is then saved into a file, inserted as a database record, or transmitted to some other geospatial service. Source data can also be captured using a GPS receiver and exported into a format for another program, or accessed in real time by an application.&lt;/p&gt;  &lt;p&gt;In its simplest form information can be made digital (that is, digitized) by scanning photographs or hard copy maps into digital image files. Or, for locational point data, coordinates can be saved into a text file and used by many open source geospatial applications.&lt;/p&gt;  &lt;p&gt;Desktop mapping programs often include the ability to draw shapes on top of other maps. The focus of desktop mapping programs is not always data creation, so their ability to do so varies greatly.&lt;/p&gt;  &lt;h4&gt;&lt;a href=&quot;http://openev.sourceforge.net/&quot;&gt;OpenEV&lt;/a&gt;&lt;/h4&gt;  &lt;p&gt;&quot;OpenEV is a library, and reference application for viewing and analyzing raster and vector geospatial data.&quot; &lt;/p&gt;  &lt;p&gt;If you want to do a quick digitizing job, OpenEV can get you up and running easily. It doesn&#39;t have a lot of fancy editing tools, but reads in dozens of raster and vector data types, which you can then use as a base for drawing your own shapes. It has a Python scripting environment and includes many image enhancement tools.&lt;/p&gt;  &lt;h4&gt;&lt;a href=&quot;http://qgis.org/&quot;&gt;Quantum GIS&lt;/a&gt;&lt;/h4&gt;  &lt;p&gt;&quot;Quantum GIS (QGIS) is a Geographic Information System that runs on Linux, Unix, Mac OS X, and Windows. QGIS supports vector, raster, and database formats.&quot; &lt;/p&gt;  &lt;p&gt;It can access PostGIS databases, in addition to dozens of other vector and raster formats. It supports feature labeling and has a great user community. Extensibility is provided through a plugin environment.&lt;/p&gt;  &lt;h3&gt;Conversion&lt;/h3&gt;  &lt;p&gt;Many geospatial projects require significant amounts of data conversion. It is not uncommon to spend as much as 80 percent of your time converting data between formats and fine-tuning the way the data is organized. De facto data format standards (for example, ESRI shapefiles for vector data; GeoTIFF for raster/image data) can help you choose a format to use if you are flexible, but depending on the programs used in a project, a particular format may be required.&lt;/p&gt;  &lt;h4&gt;&lt;a href=&quot;http://www.gdal.org/&quot;&gt;GDAL/OGR&lt;/a&gt;&lt;/h4&gt;  &lt;p&gt;&quot;GDAL/OGR is a translator library for raster geospatial data formats that is released under an X/MIT-style open source license. As a library, it presents a single abstract data model to the calling application for all supported formats. The related OGR library (which lives within the GDAL source tree) provides a similar capability for simple features vector data.&quot; &lt;/p&gt;  &lt;p&gt;The most powerful tools for data conversion are part of the GDAL/OGR project. This project comes with several command-line utilities for converting and projecting raster and vector data. The GDAL utilities handle raster data and OGR utilities handle vector data. Both are bundled together under the GDAL project banner.&lt;/p&gt;  &lt;p&gt;GDAL/OGR libraries can be used in your own applications that require conversion or data access. The pre-made command-line utilities can also be very helpful. The &lt;a href=&quot;http://fwtools.maptools.org/&quot;&gt;FWTools package&lt;/a&gt;  makes it easy to install these tools. Try using the &lt;code&gt;gdal_translate&lt;/code&gt; or &lt;code&gt;ogr2ogr&lt;/code&gt; programs to convert between formats. For example:&lt;/p&gt;  &lt;p&gt;To convert a JPEG image into a GeoTIFF file:&lt;br /&gt;&lt;code&gt;gdal_translate input.jpg output.tif&lt;/code&gt;&lt;/p&gt;  &lt;p&gt;To convert an ESRI Shapefile into GML:&lt;br /&gt;&lt;code&gt;ogr2ogr -f &quot;GML&quot; output.gml input.shp&lt;/code&gt;&lt;/p&gt;  &lt;h4&gt;&lt;a href=&quot;http://avce00.maptools.org/avce00/index.html&quot;&gt;AVCE00&lt;/a&gt;&lt;/h4&gt;  &lt;p&gt;&quot;AVCE00 is an ... ANSI-C library that makes Arc/Info (binary) Vector Coverages appear as E00! It allows you to read and write binary coverages just as if they were E00 files. ... For those who do not need a library but simply want to convert some coverages, the package includes the AVCIMPORT and AVCEXPORT conversion programs.&quot; &lt;/p&gt;  &lt;p&gt;This is a handy tool because many people think they need to have Arc/Info to convert E00 export files into coverages. Arc/Info export files are one of the formats not supported by GDAL/OGR. You can use the &lt;code&gt;avcimport&lt;/code&gt; program to import E00 files and create binary coverages with them. Then OGR utilities can be used to access or convert the coverages into other formats.&lt;/p&gt;  &lt;h3&gt;Manipulation&lt;/h3&gt;  &lt;p&gt;Of course, data is not always ready to use, even if it has been properly converted. Data often requires manipulation.  There are many types of manipulation that may be needed, such as removing unwanted features, adding fields, changing attribute values, clipping features with other features, creating buffered polygons from a line, and so on.&lt;/p&gt;  &lt;p&gt;There are two sets of tools that cover a large portion of geospatial data manipulation needs. Both can use the extended power of the GEOS libraries for advanced geometric operations.&lt;/p&gt;  &lt;h4&gt;&lt;a href=&quot;http://www.gdal.org/&quot;&gt;GDAL/OGR&lt;/a&gt;&lt;/h4&gt;  &lt;p&gt;The GDAL/OGR command-line utilities are not just good for converting data, but can also manipulate raster and vector datasets. The &lt;code&gt;gdal_translate&lt;/code&gt; and &lt;code&gt;ogr2ogr&lt;/code&gt; utilities have several options for selecting subsets of information or making changes in the output files. They can also be used to project data into particular spatial reference systems (AKA projections). For example:&lt;/p&gt;  &lt;p&gt;Use an SQL-like &lt;code&gt;SELECT&lt;/code&gt; statement to only output certain features into a new file:&lt;br /&gt;&lt;code&gt;ogr2ogr -select &quot;county = &#39;Oakland&#39;&quot; output.shp input.shp&lt;/code&gt;.&lt;/p&gt;  &lt;p&gt;Clip an image to only output the portion of an image within a certain geographic area:&lt;br /&gt;&lt;code&gt;gdal_translate -projwin -122 45 -120 55 input.tif output.tif&lt;/code&gt;&lt;/p&gt;  &lt;h4&gt;&lt;a href=&quot;http://postgis.refractions.net/&quot;&gt;PostGIS&lt;/a&gt;&lt;/h4&gt;  &lt;p&gt;Conversion tools like &lt;code&gt;ogr2ogr&lt;/code&gt; can take vector data files and export them into a PostGIS-enabled database. Because PostGIS is a PostgreSQL database that can also store geometric (spatial) data types, it is an excellent way to bring tabular and spatial data together into a common management environment. &lt;/p&gt;  &lt;p&gt;PostGIS has the ability to manipulate data as well as store it. This provides GIS-like abilities within an SQL database environment. The SQL functions include &lt;code&gt;buffer&lt;/code&gt;, &lt;code&gt;intersection&lt;/code&gt;, &lt;code&gt;within&lt;/code&gt;, &lt;code&gt;distance&lt;/code&gt;, and more. These functions take geometric data from columns in PostGIS tables and return new geometries or other information. For example, the &lt;code&gt;distance&lt;/code&gt; function will compute the distance between spatial features, and the &lt;code&gt;buffer&lt;/code&gt; function will return a new geometry that is a polygon buffered at a certain distance from the source feature. &lt;/p&gt;  &lt;p&gt;Here is an example of an SQL query that selects a particular point feature and applies a radial buffer of 100 meters to create a polygon:&lt;/p&gt;  &lt;pre&gt;&lt;code&gt;&lt;br /&gt;SELECT buffer(geo_data,100)&lt;br /&gt;FROM city_points;&lt;br /&gt;&lt;/code&gt;&lt;/pre&gt;  &lt;h4&gt;&lt;a href=&quot;http://geos.refractions.net/&quot;&gt;Geometry Engine Open Source (GEOS)&lt;/a&gt;&lt;/h4&gt;  &lt;p&gt;If you are writing your own applications, particularly in C++, you can use GEOS libraries to give you spatial manipulation capabilities. Both GDAL and PostGIS can use GEOS to allow advanced capabilities of feature manipulation.  Some of the PostGIS functions mentioned above require the GEOS libraries. GDAL can use GEOS functions if you use GDAL in a programming environment.  &lt;/p&gt;  &lt;h3&gt;Visualization&lt;/h3&gt;  &lt;p&gt;Projects that have a mapping component need some sort of visual output. The output could be a graphic file or paper printout.&lt;/p&gt;  &lt;p&gt;All the applications mentioned under the &lt;i&gt;Creation&lt;/i&gt;  section above have the ability to draw and color map data. There are many other options, some of which also include editing functionality. Although there are desktop mapping environments available, interactive web mapping is certainly the topic of the day.&lt;/p&gt;  &lt;h4&gt;&lt;a href=&quot;http://mapserver.gis.umn.edu/&quot;&gt;MapServer&lt;/a&gt;&lt;/h4&gt;  &lt;p&gt;This popular internet mapping application has both programming interfaces and a web CGI mode. Programming interfaces exist for several languages.  It also includes command-line tools for creating static image files of maps, legends, scale bars, and so on.&lt;/p&gt;  &lt;p&gt;There are a couple of &lt;em&gt;frameworks&lt;/em&gt; for building MapServer applications using PHP. Consider using one of these to get started quickly:&lt;/p&gt;  &lt;ul&gt;&lt;li&gt;&lt;a href=&quot;http://www.mapbender.org/&quot;&gt;MapBender&lt;/a&gt;&lt;/li&gt;&lt;li&gt;&lt;a href=&quot;http://chameleon.maptools.org/&quot;&gt;Chameleon&lt;/a&gt;&lt;/li&gt;&lt;/ul&gt;   &lt;h3&gt;Other Applications&lt;/h3&gt;  &lt;p&gt;Many people are using other applications to meet their daily needs. Here are some more options.&lt;/p&gt;  &lt;h4&gt;&lt;a href=&quot;http://grass.itc.it/&quot;&gt;GRASS GIS&lt;/a&gt;&lt;/h4&gt;  &lt;p&gt;&quot;The Geographic Resources Analysis Support System, commonly referred to as GRASS GIS, is a Geographic Information System  used for data management, image processing, graphics production, spatial modelling, and visualization of many types of data.&quot;  &lt;/p&gt;  &lt;p&gt;This full-fledged, desktop GIS has been used for many years by the government, academia, and industry to do sophisticated analysis and mapping.&lt;/p&gt;  &lt;h4&gt;&lt;a href=&quot;http://openmap.bbn.com/&quot;&gt;OpenMap&lt;/a&gt;&lt;/h4&gt;  &lt;p&gt;&quot;Using OpenMap, you can quickly build applications and applets that access data from legacy databases and applications. OpenMap provides the means to allow users to see and manipulate geospatial information.&quot;&lt;/p&gt;  &lt;p&gt;This JavaBean-based toolkit includes many of the standard desktop GIS features, including visualization and editing.&lt;/p&gt;  &lt;h4&gt;&lt;a href=&quot;http://gmt.soest.hawaii.edu/&quot;&gt;Generic Mapping Tools (GMT)&lt;/a&gt;&lt;/h4&gt;  &lt;p&gt;&quot;GMT is an open source collection of 60 tools for manipulating geographic and Cartesian data sets (including filtering, trend fitting, gridding, projecting, and so on) and producing Encapsulated PostScript File (EPS) illustrations ranging from simple x-y plots via contour maps to artificially illuminated surfaces and 3-D perspective views.&quot; &lt;/p&gt;  &lt;p&gt;These are command-line tools that use text configuration files, and are largely used by the oceanographic science community.&lt;/p&gt;  &lt;h4&gt;&lt;a href=&quot;http://udig.refractions.net/&quot;&gt;uDig&lt;/a&gt;&lt;/h4&gt;  &lt;p&gt;&quot;User-friendly Desktop Internet GIS (uDig) is an open source spatial data viewer/editor, with special emphasis on the OpenGIS standards for internet GIS, the Web Map Server and Web Feature Server standards. uDig will provide a common Java platform for building spatial applications with open source components.&quot;&lt;/p&gt;  &lt;p&gt;This is a Java (Eclipse)-based application with many features, and is highly extensible.&lt;/p&gt;  &lt;h4&gt;&lt;a href=&quot;http://jump-project.org/&quot;&gt;JUMP&lt;/a&gt;&lt;/h4&gt;  &lt;p&gt;&quot;The JUMP Unified Mapping Platform (JUMP) is a GUI-based application for viewing and processing spatial data. It includes many functions common to other popular GIS products for the analysis and manipulation of geospatial data.&quot; &lt;/p&gt;  &lt;p&gt;This  feature-rich desktop GIS environment can create, edit, and manipulate data. It can do much more through additional plugins. JUMP is a Java application and has several levels of programming interfaces available.&lt;/p&gt;  &lt;h3&gt;Conclusion&lt;/h3&gt;  &lt;p&gt;Open source geospatial applications and programming environments can fill all of the standard components of a geospatial project.  The geospatial landscape is becoming rich with choice.&lt;/p&gt;  &lt;p&gt;&lt;i&gt;Please note: the quoted material for each tool discussed in this article comes from that tool&#39;s specific web site. &lt;/i&gt;&lt;/p&gt;  &lt;p&gt;&lt;i&gt; &lt;a href=&quot;http://www.oreillynet.com/pub/au/1898&quot;&gt;Tyler Mitchell&lt;/a&gt;  is the author of &lt;a href=&quot;http://www.oreilly.com/catalog/webmapping/&quot;&gt;Web Mapping Illustrated&lt;/a&gt; - a book focused on teaching how to use popular Open Source Geospatial Toolkits.  He works as the Executive Director of the &lt;a href=&quot;http://osgeo.org/&quot; target=&quot;_blank&quot;&gt;Open Source Geospatial Foundation&lt;/a&gt;, aka OSGeo.&lt;/i&gt;&lt;/p&gt;&lt;p&gt;&lt;i&gt;&lt;br /&gt;&lt;/i&gt;&lt;/p&gt;&lt;/span&gt;Source: &lt;a href=&quot;http://www.oreillynet.com/pub/a/network/2005/06/10/osgeospatial.html&quot;&gt;O&#39;Reilly Network -- An Introduction to Open Source Geospatial Tools&lt;/a&gt;&lt;br /&gt;.</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/359163709543769689/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/359163709543769689' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/359163709543769689'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/359163709543769689'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2008/07/introduction-to-open-source-geospatial.html' title='An Introduction to Open Source Geospatial Tools'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-7491715878677777828</id><published>2008-07-10T21:52:00.002+07:00</published><updated>2008-07-10T22:02:33.114+07:00</updated><title type='text'>Bayesian Network and AHP</title><content type='html'>บังเอิญผมนั่งหา paper ดูว่ามีใครทำ Bayesian Network โดยใช้ Python แล้วก็ ArcGIS บ้างเพราะผมเองยังมีความคิดอยากจะรันโมเดลทั้งหมดโดยใช้ Model Builder ใน ArcGIS อยู่เลย&lt;br /&gt;เหตุผลก็ง่ายๆเลยครับ หึหึ เพราะ&lt;br /&gt;1. ผมพอใช้ ArcGIS เป็นถึงระดับหนึ่ง นี่เป็นเหตุหลักเลยที่ไม่ต้องไปนั่งทำความคุ้นเคยอื่นกับโปรแกรมอย่างพวก Cube หรือ GRASS มันอาจช่วยให้ผมจบไวขึ้น&lt;br /&gt;2. ArcGIS นี่มี Model Builder เป็นลักษณะการสร้าง model เป็น input-output คล้ายกับ Cube เลย&lt;br /&gt;3. ArcGIS มี Geoprocessing language แล้วก็ network analyst model ส่วนพวกนี้อาจช่วยในการวิเคราะห์โมเดลในส่วนอื่นๆได้ในอนาคต&lt;br /&gt;4. ArcGIS สร้าง custom model ได้โดยใช้ python ภาษาในฝันของผม(แต่ยังใช้ไม่เป็น)&lt;br /&gt;โดยประการฉะนี้ก็เลยพยายามจะไปให้ถึงดวงดาวโดยการใช้&lt;br /&gt;ArcGIS, Python, BayesianNetwork Library (อย่างเช่น Mocapy) ในการทำ thesis แต่ไม่รู้จะเป็นไงบ้าง&lt;br /&gt;&lt;br /&gt;พูดซะยาวบังเอิญไปเจอ paper Bayesian Network กับ AHP และ ก็ GIS นิดหน่อยในการช่วยวิเคราะห์แหล่งท่องเที่ยวครับน่าสนใจดี Paper ต้องโหลดใน Science Direct ครับตามนี้&lt;br /&gt;&lt;br /&gt;&lt;a href=&quot;http://www.sciencedirect.com/science?_ob=ArticleURL&amp;amp;_udi=B6V03-4R33560-2&amp;amp;_user=578664&amp;amp;_coverDate=11%2F07%2F2007&amp;amp;_alid=765173190&amp;amp;_rdoc=1&amp;amp;_fmt=high&amp;amp;_orig=search&amp;amp;_cdi=5635&amp;amp;_sort=d&amp;amp;_docanchor=&amp;amp;view=c&amp;amp;_ct=1&amp;amp;_acct=C000029478&amp;amp;_version=1&amp;amp;_urlVersion=0&amp;amp;_userid=578664&amp;amp;md5=a84f9c0a81a8a2072a4dac2a14f33e0f&quot;&gt;&lt;span style=&quot;font-weight: bold;&quot;&gt;A&lt;span style=&quot;color: rgb(255, 0, 0); font-weight: bold;&quot;&gt; Bayesian network&lt;/span&gt; and analytic hierarchy process based personalized recommendations for tourist attractions over the Internet&lt;/span&gt;&lt;/a&gt;&lt;br /&gt;&lt;i&gt;Expert Systems with Applications&lt;/i&gt;, &lt;i&gt;&lt;b&gt;&lt;span style=&quot;color:#ff0000;&quot;&gt;In Press, Corrected Proof&lt;/span&gt;&lt;/b&gt;&lt;/i&gt;, &lt;i&gt;Available online 7 November 2007&lt;/i&gt;&lt;br /&gt;Yuxia Huang, Ling Bian</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/7491715878677777828/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/7491715878677777828' title='1 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/7491715878677777828'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/7491715878677777828'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2008/07/bayesian-network-and-ahp.html' title='Bayesian Network and AHP'/><author><name>คิม</name><uri>http://www.blogger.com/profile/09715683525677301971</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='31' height='22' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiVqiQH38LYmJPmJm4ByuUPBTd-azQRs5CBS6wJOLsasghtB5GgBSfVdMUfdlULQzJKmB8bi-FIIQju3qji3w03sypZX7FViKw1CLbR85wMUY76AExFnpTyS5dIDXrotw/s220/551000003897601.jpg'/></author><thr:total>1</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-1289699239421824029</id><published>2008-06-22T22:40:00.002+07:00</published><updated>2008-06-22T22:46:39.474+07:00</updated><category scheme="http://www.blogger.com/atom/ns#" term="microscopic simulation"/><title type='text'>Calibrating Large Scale Microscopic Simulation: Recent Pulication</title><content type='html'>This recent paper might be a useful reference &lt;br /&gt;&lt;a href=&quot;http://scitation.aip.org/getabs/servlet/GetabsServlet?prog=normal&amp;id=JTPEDI000134000007000273000001&amp;idtype=cvips&amp;gifs=Yes&amp;type=ALERT&quot;&gt;&lt;br /&gt;Large-Scale Microscopic Simulation: Toward an Increased Resolution of Transportation Models&lt;/a&gt;&lt;br /&gt;&lt;br /&gt;J. Transp. Engrg., Volume 134, Issue 7, pp. 273-281 (July 2008)&lt;br /&gt;&lt;br /&gt;Mark C. Smith,1 P.E.; Adel W. Sadek,2 A.M.ASCE; and Shan Huang3&lt;br /&gt;1Research Assistant, School of Engineering, Civil and Environmental Engineering Program, Univ. of Vermont, Burlington, VT 05405. E-mail: msmith@uvm.edu&lt;br /&gt;2Associate Professor, School of Engineering, Civil and Environmental Engineering Program, Univ. of Vermont, Burlington, VT 05405 (corresponding author). E-mail: asadek@cems.uvm.edu&lt;br /&gt;3Research Assistant, School of Engineering, Civil and Environmental Engineering Program, Univ. of Vermont, Burlington, VT 05405. E-mail: shuang@uvm.edu&lt;br /&gt;&lt;br /&gt;(Accepted 26 December 2007)&lt;br /&gt;&lt;br /&gt;Recent years have witnessed an interest in increasing the resolution of transportation models to allow them to better address emerging transportation issues, such as concerns about emissions, air quality, and emergency preparedness. Microscopic simulation models, especially those that combine routing logic with accurate modeling of traffic flow dynamics, have vast untapped potential for modeling large-scale transportation networks and for achieving the resolution level required for addressing such emerging issues. Building and calibrating such models, however, is quite challenging and is currently not widely understood. This paper describes the process of developing and calibrating a large-scale microscopic model of Chittenden County, Vt., an area of about 1,400  km2 (540  mi2). The model was developed primarily using readily available data to which most metropolitan planning organizations are expected to have access. Following a brief description of how the network was coded and the traffic demand specified, the paper discusses the process of error checking and model calibration. Preliminary calibration results are encouraging, given the complexity of the model. The paper includes a summary of the main lessons learned and modeling pitfalls that should be avoided.&lt;br /&gt;&lt;br /&gt;©2008 ASCE</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/1289699239421824029/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/1289699239421824029' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/1289699239421824029'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/1289699239421824029'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2008/06/calibrating-large-scale-microscopic.html' title='Calibrating Large Scale Microscopic Simulation: Recent Pulication'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-602194778771776430</id><published>2008-02-02T02:14:00.000+07:00</published><updated>2008-02-02T03:50:47.646+07:00</updated><category scheme="http://www.blogger.com/atom/ns#" term="activity based"/><category scheme="http://www.blogger.com/atom/ns#" term="calibration"/><category scheme="http://www.blogger.com/atom/ns#" term="dynamic traffic assignment"/><category scheme="http://www.blogger.com/atom/ns#" term="integration"/><title type='text'>Approaches to Fine Tune the Acitivity Based Model</title><content type='html'>&lt;span&gt;&lt;span&gt;&lt;span&gt;&lt;span&gt;&lt;span&gt;จำได้ว่าเคยถกกับคิมเรื่องทำอย่างไรจะตรวจสอบความถูกต้อง (validate)&lt;br /&gt;รวมถึงปรับแก้ (calibrate) แบบจำลองแบบ activity based ที่กำลังจะพัฒนาขึ้น&lt;br /&gt;เรื่องนี้ผมเคยเสนอว่ามีแนวทางการตรวจสอบอยู่สองทาง&lt;br /&gt;&lt;br /&gt;วิธีแรกได้แก่การเก็บข้อมูลภาคสนามโดยเก็บกิจกรรมที่ทำ&lt;br /&gt;เพื่อการตรวจสอบความถูกต้องโดยเฉพาะ&lt;br /&gt;อย่างไรก็ตามเป็นวิธีที่ค่อนข้างจะลำบากอยู่สักหน่อยเนื่องจากต้องเสียค่าใช้จ่ายเพิ่มเติม&lt;br /&gt;และไม่แน่ว่าจะใช้ตรวจสอบได้ดีหรือไม่&lt;br /&gt;เพราะว่ากิจกรรมในแต่ละวันของแต่ละบุคคลนั้นค่อนข้างหลากหลาย&lt;br /&gt;และอาจต้องเก็บตัวอย่างจำนวนมากพอเพื่อตรวจสอบรูปแบบการทำกิจกรรม&lt;br /&gt;ในระดับภาพรวม (macro)&lt;br /&gt;&lt;br /&gt;วิธีที่สองเป็นการเก็บข้อมูลข้างเคียงที่เป็นผลพลอยได้จากการทำกิจกรรมหรือการเดินทาง&lt;br /&gt;เช่นข้อมูลจากผู้ใช้บริการขนส่งมวลชน ข้อมูลการเดินทางจากแบบจำลอง 4 steps&lt;br /&gt;โดยเฉพาะอย่างยิ่งข้อมูลการไหลของจราจร (link flow) ที่เกิดขึ้นจริง&lt;br /&gt;อย่างไรก็ตามในขณะนั้นยังไม่มีไอเดียเพิ่มเติมว่าหากค่าที่ได้จากแบบจำลอง&lt;br /&gt;คลาดเคลื่อนไปจากค่าที่สังเกตุได้จริงแล้วจะปรับแก้ในส่วนของ&lt;br /&gt;activity based model กันอย่างไร&lt;br /&gt;&lt;br /&gt;เมื่อเร็วๆนี้ Lin et al. (2007) ซึ่งรวมถึง Bhat (คาดว่าอยู่แล็บเดียวกัน)&lt;br /&gt;ได้เสนอแนวคิดในการบูรณาการระหว่าง &lt;br /&gt;Activity Based Model (ที่เขามองว่าเป็น Demand Side) กับ&lt;br /&gt;Dynamic Traffic Assignment (DTA)  (ที่เขามองว่าเป็น Supply Side หรือเราอาจมองว่ามาจาก Trip Based Model) เข้าด้วยกัน&lt;br /&gt;ถ้าผมเข้าใจไม่ผิด นั่นคือแนวคิดในการปรับแก้ Activity Based Model โดยใช้ Link Flow ผ่านทาง DTA&lt;br /&gt;ตัว paper&lt;br /&gt;&quot;&lt;a href=&quot;http://pubsindex.trb.org/document/view/default.asp?lbid=848615&quot;&gt;Integration of Activity-Based Modeling and Dynamic Traffic Assignment&lt;/a&gt;&quot;&lt;br /&gt;&lt;br /&gt;แต่อย่างไรก็ตามเนื่องจากมันเป็นการ integrate สองแบบจำลองเข้าด้วยกัน&lt;br /&gt;มันถูกอาจมองว่าเป็นการปรับแก้ DTA โดยใช้ Activity Based ก็ได้?&lt;br /&gt;(คือผลลัพธ์มันอาจจะอยู่ตรงกลางๆ)&lt;br /&gt;&lt;br /&gt;ยังไม่ได้อ่าน ตัวpaper ฉบับเต็ม ผิดพลาดอย่างไรต้องขออภัย&lt;br /&gt;คิมช่วยเอาฉบับเต็มมาอ่านหน่อยครับ&lt;br /&gt;&lt;/span&gt;&lt;/span&gt;&lt;/span&gt;&lt;/span&gt;&lt;/span&gt;&lt;span&gt;&lt;br /&gt;&lt;/span&gt;&lt;span&gt;้&lt;/span&gt;</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/602194778771776430/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/602194778771776430' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/602194778771776430'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/602194778771776430'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2008/02/approaches-to-fine-tune-acitivity-based.html' title='Approaches to Fine Tune the Acitivity Based Model'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-8318394069792062554</id><published>2007-07-18T10:18:00.000+07:00</published><updated>2007-07-18T10:36:19.523+07:00</updated><category scheme="http://www.blogger.com/atom/ns#" term="writing"/><title type='text'>Use of the word &#39;Data&#39;</title><content type='html'>&lt;span&gt;Data is information, usually in the form of facts or statistics that can be analysed or used in further calculations.&lt;br /&gt;&lt;/span&gt;&lt;i&gt;&lt;span&gt;Such tasks require the worker to process a large amount of data.&lt;br /&gt;  This will make the data easier to collect.&lt;/span&gt;&lt;/i&gt;&lt;span&gt;&lt;br /&gt;&lt;br /&gt;Data is usually regarded as an uncount noun and is used with a singular form of a verb.&lt;br /&gt;&lt;/span&gt;&lt;i&gt;&lt;span&gt;The data was still being processed at the Census Office.&lt;br /&gt;  ...whenever the data involves confrontation between nuclear powers.&lt;/span&gt;&lt;/i&gt;&lt;span&gt;&lt;br /&gt;&lt;br /&gt;People usually say this data, rather than `these data&#39;.&lt;br /&gt;&lt;/span&gt;&lt;i&gt;&lt;span&gt;Processing this data only takes a moment.&lt;br /&gt;  He may be incapable of transferring this data correctly to a patient&#39;s records.&lt;/span&gt;&lt;/i&gt;&lt;span&gt;&lt;br /&gt;&lt;br /&gt;However, some people think these uses are incorrect. They say that data is the plural form of the noun `datum&#39;, and should therefore be used with a plural form of a verb. They also say that you should talk about these data, not `this data&#39;.&lt;br /&gt;&lt;/span&gt;&lt;i&gt;&lt;span&gt;He now denies that all knowledge comes from the senses at the moment the sense data are received.&lt;br /&gt;  For the most part these data are properly handled by conventional dualistic logic.&lt;/span&gt;&lt;/i&gt;&lt;span&gt;&lt;br /&gt;&lt;br /&gt;&lt;/span&gt;&lt;span&gt;&lt;span&gt;&lt;strong&gt;It is probably best in any kind of formal or scientific writing to use a plural form of a verb with data and to talk about these data rather than `this data&#39;.&lt;/strong&gt;&lt;/span&gt;&lt;/span&gt;&lt;span&gt;&lt;span&gt; In other situations, you can use either form.&lt;/span&gt;&lt;span&gt;&lt;strong&gt;&lt;br /&gt;&lt;/strong&gt;&lt;/span&gt;&lt;/span&gt;&lt;span&gt;&lt;strong&gt;&lt;br /&gt;&lt;/strong&gt;&lt;/span&gt;&lt;span&gt;&lt;strong&gt;Source:&lt;br /&gt;(c) HarperCollins Publishers.&lt;br /&gt;&lt;br /&gt;&lt;br /&gt;&lt;/strong&gt;&lt;/span&gt;</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/8318394069792062554/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/8318394069792062554' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/8318394069792062554'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/8318394069792062554'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2007/07/using-of-word-data-data-is-information.html' title='Use of the word &#39;Data&#39;'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><thr:total>0</thr:total></entry><entry><id>tag:blogger.com,1999:blog-2451586136929499042.post-8172470281562695067</id><published>2007-07-11T16:21:00.000+07:00</published><updated>2007-07-18T10:29:01.077+07:00</updated><category scheme="http://www.blogger.com/atom/ns#" term="talk"/><title type='text'>Hello World !!</title><content type='html'>This is the first post at Transiit.</content><link rel='replies' type='application/atom+xml' href='http://transiit.blogspot.com/feeds/8172470281562695067/comments/default' title='Post Comments'/><link rel='replies' type='text/html' href='http://www.blogger.com/comment/fullpage/post/2451586136929499042/8172470281562695067' title='0 Comments'/><link rel='edit' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/8172470281562695067'/><link rel='self' type='application/atom+xml' href='http://www.blogger.com/feeds/2451586136929499042/posts/default/8172470281562695067'/><link rel='alternate' type='text/html' href='http://transiit.blogspot.com/2007/07/hello-world.html' title='Hello World !!'/><author><name>MP</name><uri>http://www.blogger.com/profile/10136520375549053590</uri><email>noreply@blogger.com</email><gd:image rel='http://schemas.google.com/g/2005#thumbnail' width='32' height='32' src='//blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjTFMRNNHUSGPaCKheMeezX6G2aowEf6zhoNExwQZ2Jny8Sw6iDTVATtToeN4dZbGgvxVQZ4j9nl82roWZ26QoEM6GqrV00IERr4UeTMZ4k1v_oweZSU8f-r0ZTcB7yfPM/s220/Mongkut.jpg'/></author><thr:total>0</thr:total></entry></feed>