{"id":8440,"date":"2015-12-15T14:12:01","date_gmt":"2015-12-15T06:12:01","guid":{"rendered":"https:\/\/ihower.tw\/blog\/?p=8440"},"modified":"2015-12-15T14:16:22","modified_gmt":"2015-12-15T06:16:22","slug":"%e5%8f%b0%e7%81%a3%e8%b3%87%e6%96%99%e7%a7%91%e5%ad%b8%e6%84%9b%e5%a5%bd%e8%80%85%e5%b9%b4%e6%9c%83-2015-%e5%bf%83%e5%be%97%e7%ad%86%e8%a8%98","status":"publish","type":"post","link":"https:\/\/ihower.tw\/blog\/8440-%e5%8f%b0%e7%81%a3%e8%b3%87%e6%96%99%e7%a7%91%e5%ad%b8%e6%84%9b%e5%a5%bd%e8%80%85%e5%b9%b4%e6%9c%83-2015-%e5%bf%83%e5%be%97%e7%ad%86%e8%a8%98","title":{"rendered":"\u53f0\u7063\u8cc7\u6599\u79d1\u5b78\u611b\u597d\u8005\u5e74\u6703 2015 \u5fc3\u5f97\u7b46\u8a18"},"content":{"rendered":"<p>\u4eca\u5e74 2015\/8\/22-23 \u53bb\u53c3\u52a0\u4e86\u53f0\u7063\u8cc7\u6599\u79d1\u5b78\u611b\u597d\u8005\u5e74\u6703\uff0c\u9019\u662f\u7576\u6642\u7684\u7b46\u8a18\u6574\u7406\u3002\u53e6\u5916\u9084\u6709 <a href=\"https:\/\/hackpad.com\/2015--g9T8HKd6TD0\">\u6703\u773e\u7248\u7684 hackpad \u7b46\u8a18<\/a><\/p>\n<p><!--more--><\/p>\n<h2 id=\"toc_0\">Keynote- \u8eca\u54c1\u89ba \u9999\u6e2f\u4eba\u3001\u963f\u91cc\u5df4\u5df4 \u6578\u64da\u526f\u7e3d\u88c1<\/h2>\n<ul>\n<li>\u81ea\u7a31\u4e0d\u662f\u8cc7\u6599\u79d1\u5b78\u5bb6\uff0c\u800c\u662f\u559c\u6b61\u7528\u6578\u64da\u7684\u7522\u54c1\u7d93\u7406<\/li>\n<li>\u4e0d\u662f\u6240\u6709\u554f\u984c\u90fd\u53ef\u4ee5\u7528\u6578\u64da\u89e3\u6c7a<\/li>\n<li>\u4e00\u5207\u5f9e\u5b9a\u7fa9\u554f\u984c\u958b\u59cb\uff0c\u4ee5\u77e5\u5230\u672a\u77e5\u7684\u904e\u7a0b<\/li>\n<li>\u5c0f\u5fc3 unexpected \u7684\u8cc7\u6599\uff0c\u8981\u53bb\u6536\u96c6\u8ca0\u9762\u6578\u64da\uff0c\u4f8b\u5982\u9664\u4e86\u7d00\u9304\u5403\u4e86\u54ea\u88e1\uff0c\u4e5f\u7d00\u9304\u6c92\u5403\u5b8c\u54ea\u4e9b<\/li>\n<li>\u62db\u52df\u5df2\u7d93\u62c6\u6210\u5169\u7a2e: data scientist\u3001decision scientist<\/li>\n<li>\u5047\u8a2d\u6578\u64da\u90fd\u80fd\u7372\u53d6\u53bb\u601d\u8003\u554f\u984c\uff0c\u5305\u62ec\u884d\u751f\u53ca\u8ff4\u8def:    \u4f8b\u5982 uber\uff0c\u5728\u4ee5\u524d\u6c92\u8fa6\u6cd5\u601d\u8003\u8a08\u7a0b\u8eca\u8f09\u5ba2\u6578\u64da<\/li>\n<li>\u7528\u63a2\u91dd\u8a2d\u5099\u53ef\u4ee5\u6536\u96c6\u6703\u5834\u7684\u6d41\u91cf\uff0c\u5c31\u4e0d\u9700\u8981\u7528\u554f\u5377\u4e86: \u4e00\u7a2e IoT<\/li>\n<li>\u4ed6\u554f\u8001\u95c6: \u4f5c\u70ba data leader\uff0c\u6211\u8981\u4f5c\u4ec0\u9ebc\u8b93\u4f60\u723d? \u7e3d\u88c1\u8aaa: \u6211\u60f3\u77e5\u9053\u6211\u4e0a\u500b\u79ae\u62dc\u505a\u4e86\u4ec0\u9ebc right? \u4ec0\u9ebc wrong?<\/li>\n<li>\u66f4\u5927\u7684\u554f\u984c\u662f\u6578\u64da\u5931\u7f3a\u7684\u554f\u984c: \u771f\u6b63\u8981\u81ea\u52d5\u5316\uff0c\u9700\u8981\u66f4\u7a69\u5b9a\u7684\u591a\u91cd model\uff0c\u800c\u4e0d\u53ea\u662f\u5be6\u9a57\u5831\u544a<\/li>\n<li>\u8981\u628a data availability \u554f\u984c\u7576\u4f5c\u5de5\u7a0b SLA \u554f\u984c\uff0c\u9019\u624d\u662f\u5de5\u7a0b\u5316\u5168\u9762\u670d\u52d9\uff0c\u800c\u4e0d\u662f\u79d1\u5b78\u5bb6\u7684 toy<\/li>\n<li>\u8b93\u6578\u64da\u5f9e support \u5230\u7522\u751f\u884c\u52d5 -&gt; feedback &gt; \u6539\u8b8a\u4e0b\u6b21\u6c7a\u7b56 <\/li>\n<li>\u8b93\u6578\u64da\u53bb\u52a9\u529b\u63cf\u8ff0\u3001\u8a3a\u65b7\u3001\u9810\u6e2c\u3001\u884c\u52d5\u5efa\u8b70<\/li>\n<li>\u8981\u5c0d\u6578\u64da\u7576\u4f5c\u8cc7\u7522\uff0c\u624d\u80fd\u5920\u4f5c\u5927\u6578\u64da\u884c\u696d<\/li>\n<li>\u624b\u6a5f\u662f\u6578\u64da\u5206\u6790\u7684\u6700\u5927\u7684\u6578\u64da\u6536\u96c6\u611f\u77e5\u5668\u8ddf\u5a92\u9ad4<\/li>\n<li>data product \u6703\u53bb\u672a\u4f86\u7684\u5e38\u614b\uff0c\u4f8b\u5982 smart \u6c34\u676f\u7d00\u9304\u559d\u4e86\u591a\u5c11\u6c34\uff0c\u7136\u5f8c\u50b3\u5230\u624b\u6a5f\u4e0a<\/li>\n<li>\u5feb+\u6e96\u7684\u6578\u64da: \u5f9e\u5df2\u77e5\u898f\u5f8b\u4e2d\u7522\u751f\u50f9\u503c<\/li>\n<li>\u5ee3+\u4e82\u7684\u6578\u64da: \u5f9e\u767c\u73fe\u4e2d\u985b\u8986\u5df2\u77e5<\/li>\n<li>\u6578\u64da\u7684\u9589\u74b0 (lifecycle?): \u5e03\u9ede\/\u6536\u96c6 -&gt; \u5b58\u5132\/\u5237\u65b0 -&gt; \u8fa8\u8b58\/\u95dc\u806f (\u6574\u5408) -&gt; \u6316\u6398\/\u6c7a\u7b56 -&gt; \u884c\u52d5\/\u53cd\u6f70 (\u898f\u5f8b)<\/li>\n<li>\u8fa8\u8b58\u7684\u5c0f\u6545\u4e8b: \u4f8b\u5982\u5e33\u865f\u7d66\u8001\u5a46\u7528\uff0c\u8cc7\u6599\u5beb\u7537\u751f\uff0c\u4f46\u662f\u8cfc\u7269\u884c\u70ba\u662f\u5973\u751f<\/li>\n<li>data science \u8981\u601d\u8003\u5168\u5c40: \u524d\u9762\u7684\u4eba(engineer)\u662f\u82e6\u903c\u7684\uff0c\u7e3d\u6709\u4e00\u5929\u6703\u51fa\u982d\u3002\u6c92\u6709\u524d\u65b9\uff0c\u5c31\u6c92\u6709\u5f8c\u65b9 (\u9019\u88e1\u5e6b\u5fd9\u5e73\u53cd)<\/li>\n<li>\u5927\u6578\u64da\u6280\u8853\u5c31\u662f\u52a0\u901f\u8207\u7d2f\u7a4d(\u6578\u64da\u3001\u5206\u6790\u3001\u670d\u52d9)\u7684\u80fd\u529b<\/li>\n<li>\u5927\u6578\u64da\u7684\u80fd\u529b\u6bd4\u55bb:\n<ul>\n<li>\u601d\u60f3: model\u3001\u9810\u6e2c\u3001\u79d1\u5b78\u3001\u5206\u6790\u5e2b<\/li>\n<li>\u8840\u6db2: ETL<\/li>\n<li>\u9aa8\u9abc: \u96f2, \u57fa\u672c\u8a2d\u65bd<\/li>\n<\/ul>\n<\/li>\n<li>\u79d1\u5b78\u5316 -&gt; \u6578\u64da\u5316(?) -&gt; \u7522\u54c1\u5316 -&gt; \u5e02\u5834\u5316<\/li>\n<li>\u5927\u6578\u64da\u662f\u4f86\u81ea\u5f88\u591a\u5c0f\u6578\u64da\u7684\u7d44\u5408<\/li>\n<\/ul>\n<blockquote><p>\n\u8b1b\u8005\u7528\u9ad8\u968e\u6846\u67b6\u53bb\u60f3\u6578\u64da\u8981\u5982\u4f55\u5e6b\u52a9\u6c7a\u7b56\uff0c\u6f14\u8b1b\u98a8\u683c\u6bd4\u8f03\u6c92\u6709\u6bb5\u843d\n<\/p><\/blockquote>\n<h2 id=\"toc_1\">IBM Chief Scientist<\/h2>\n<ul>\n<li>GPU vs. CPU: GPU \u7684 core \u6709\u5e7e\u5343\u500b\u662f\u8da8\u52e2\uff0c\u4f46\u662f general purpose GPU \u57fa\u790e\u9084\u6c92\u6709\u7d71\u4e00\u597d<\/li>\n<li>graph big data \u554f\u984c\u662f\u4e0b\u4e00\u500b\u5927\u6311\u6230<\/li>\n<li>\u4e0b\u4e00\u6b65\u7684\u6703\u5f80\u54ea\u88e1\u8d70&#8230; IBM chip\u2026<\/li>\n<\/ul>\n<blockquote><p>\n\u4e00\u5806 big data \u7684 buzzword\uff0c\u6295\u5f71\u7247\u98a8\u683c\u662f \u8001\u6d3e\u7814\u7a76\u6240 \u98a8\u683c\uff0c\u5716\u8868\u8d85\u8907\u96dc\n<\/p><\/blockquote>\n<h2 id=\"toc_2\">\u6e05\u5927\u7c21\u59d3\u6559\u6388<\/h2>\n<blockquote><p>\n\u807d\u4e94\u5206\u9418\u5c31\u9583\u4e86\uff0c\u6295\u5f71\u7247\u5bc6\u5bc6\u9ebb\u9ebb\u975e\u5e38\u591a\u5b57\n<\/p><\/blockquote>\n<h2 id=\"toc_3\">\u4ea4\u5927\u66fe\u59d3\u6559\u6388<\/h2>\n<blockquote><p>\n\u600e\u9ebc\u53c8 Introduction to Big Data \u4e86(\u5728\u9019\u985e\u5834\u5b50\u5df2\u7d93\u807dN\u904d\u4e86)\uff0c\u6211\u5408\u7406\u61f7\u7591\u4e3b\u8fa6\u55ae\u4f4d\u6c92\u6709\u5148\u904e\u6295\u5f71\u7247\n<\/p><\/blockquote>\n<h2 id=\"toc_4\">\u53f0\u5927\u6797\u8ed2\u7530\u8001\u5e2b\u7684 Learning for Big Data<\/h2>\n<p><iframe loading=\"lazy\" title=\"Learning for Big Data\uff0d\u6797\u8ed2\u7530\" src=\"https:\/\/www.slideshare.net\/slideshow\/embed_code\/key\/hc7QoeFDttKEfy\" width=\"427\" height=\"356\" frameborder=\"0\" marginwidth=\"0\" marginheight=\"0\" scrolling=\"no\" style=\"border:1px solid #CCC; border-width:1px; margin-bottom:5px; max-width: 100%;\" allowfullscreen> <\/iframe> <\/p>\n<div style=\"margin-bottom:5px\"> <strong> <a href=\"https:\/\/www.slideshare.net\/tw_dsconf\/learning-for-big-data\" title=\"Learning for Big Data\uff0d\u6797\u8ed2\u7530\" target=\"_blank\">Learning for Big Data\uff0d\u6797\u8ed2\u7530<\/a> <\/strong> from <strong><a href=\"https:\/\/www.slideshare.net\/tw_dsconf\" target=\"_blank\">\u53f0\u7063\u8cc7\u6599\u79d1\u5b78\u5e74\u6703<\/a><\/strong> <\/div>\n<ul>\n<li>Q1: \u554f\u984c\u600e\u9ebc\u554f? \u5f9e\u52d5\u6a5f\u548c\u80fd\u529b\u8003\u616e<\/li>\n<li>Q2: \u5982\u4f55\u6311 ML \u65b9\u6cd5? \u5148\u5f9e KISS \u958b\u59cb\n<ul>\n<li>1. linear model<\/li>\n<li>2. naive baye for frenency-based \u6a5f\u7387\u554f\u984c<\/li>\n<li>3.  random forests tree for non-linear<\/li>\n<\/ul>\n<\/li>\n<li>Q3: \u5982\u4f55\u6539\u5584 ML performance? data construction is more important <\/li>\n<li>feature engineering: \u628a\u4f60\u6240\u77e5\u9053\u7684 domain knowledge \u653e\u9032\u53bb feature extraction\n<ul>\n<li>\u6240\u4ee5\u8981\u5148\u5206\u6790\u8cc7\u6599<\/li>\n<li>\u5148\u4f5c\u57fa\u672c\u7684ML (linear, random forests)\u4e5f\u53ef\u4ee5\u77e5\u9053\u4e00\u4e9b\u8cc7\u8a0a<\/li>\n<\/ul>\n<\/li>\n<li>\u5f88\u591a\u7814\u7a76\u56e0\u70ba\u7528\u540c\u6a23\u7684 dataset\uff0c\u6240\u4ee5\u4e5f\u6709 overfitting \u554f\u984c&#8230; XD<\/li>\n<\/ul>\n<h2 id=\"toc_5\">\u6d77\u91cf\u8996\u89ba\u8cc7\u6599<\/h2>\n<ul>\n<li>\u96fb\u8166\u5716\u5b78 <a href=\"http:\/\/ditto.us.com\/\" class=\"autohyperlink\">ditto.us.com\/<\/a><\/li>\n<li><a href=\"http:\/\/www.image-net.org\/\" class=\"autohyperlink\">www.image-net.org\/<\/a> dataset \u8fa8\u8b58 object<\/li>\n<li>CNN methods(deep learning \u5f9e2012\u958b\u59cb\u5927\u5e45\u6539\u5584)<\/li>\n<li>\u4e0d\u904e\u9084\u662f\u6709 overfitting<\/li>\n<li>Place-Net \u53e6\u4e00\u500b dataset<\/li>\n<li>\u611f\u89ba\u8981\u7522\u54c1\u5316\uff0c\u5148\u5f9e\u56fa\u5b9a\u554f\u984c\u958b\u59cb\uff0c\u4f8b\u5982\u8fa8\u8b58\u51fa logo, \u5834\u666f&#8230;etc <a href=\"http:\/\/ceres-solver.org\/\" class=\"autohyperlink\">ceres-solver.org\/<\/a><\/li>\n<\/ul>\n<blockquote><p>\n\u6709\u9ede\u8b1b\u592a\u96e3\u592a\u8a73\u7d30\u4e86\uff0c\u5728\u8ac7\u8996\u89ba\u8fa8\u8b58\u6f14\u7b97\u6cd5\u4e86\u3002\u4e0d\u904e demo \u61c9\u7528\u7684\u90e8\u5206\u5f88\u6709\u8da3<br \/>\n\u8996\u89ba\u8fa8\u8b58\u5f88\u6709\u6f5b\u529b\u3001\u4f46\u662f\u53c8\u66f4\u56f0\u96e3\u4e86\n<\/p><\/blockquote>\n<ul>\n<li>Demo: \u9577\u6642\u9593\u5f71\u7247\u7684\u81ea\u52d5\u7cbe\u5f69\u526a\u8f2f\uff1a\u7531\u5f71\u7247 -&gt; \u81ea\u52d5\u62fc\u63a5\u51fa\u526a\u8f2f\u7167\u7247<\/li>\n<li>Hyperlapse<\/li>\n<li>\u7528 virtual data \u53bb train: \u4f8b\u5982\u7528\u8cfd\u8eca\u904a\u6232 train\uff0c\u7136\u5f8c\u518d\u5be6\u969b\u958b\u8eca<\/li>\n<\/ul>\n<blockquote><p>\n\u672c\u4f86\u4ee5\u70ba\u662f\u8cc7\u6599\u8996\u89ba\u5316\u7684 talk\uff0c\u6c92\u60f3\u5230\u662f\u5f71\u50cf\u3001\u5f71\u7247\u8996\u89ba\u8fa8\u8b58\u7684 Deep Learning \u61c9\u7528\uff0c\u807d\u5b8c\u771f\u662f\u5916\u661f\u79d1\u6280\u3002\n<\/p><\/blockquote>\n<h2 id=\"toc_6\">Big Education in the Era of Big Data<\/h2>\n<ul>\n<li>personalization and customization<\/li>\n<li>good \u7684\u5927\u5b78 focus \u5728\u8ad6\u6587\u6578\u91cf<\/li>\n<li>better \u7684\u5927\u5b78 focus \u5728 citation numbers and impact factors<\/li>\n<li>best \u7684\u5927\u5b78 focus on EDUCATION<\/li>\n<li>What&#39;s Big Education? \u5f9e 1vs1, 1vs100 \u5230 1vs10000<\/li>\n<li>Learning Analytics \u5b78\u7fd2\u5206\u6790<\/li>\n<li>BigScholar 2015 \u7814\u8a0e\u6703<\/li>\n<\/ul>\n<blockquote><p>\n\u611f\u89ba\u6f14\u8b1b\u5167\u5bb9\u6df1\u5ea6\u8207 &quot;\u5927\u6578\u64da&quot;\u548c&quot;Big&quot;\u7684\u51fa\u73fe\u983b\u7387\u6210\u53cd\u6bd4<br \/>\n\u807d\u5230\u73fe\u5728\u6709\u9ede\u865b\u554a&#8230;&#8230;.  \u7adf\u7136\u5728\u505a\u8da8\u52e2\u4ecb\u7d39\u548c\u4ecb\u7d39\u81ea\u5df1\u7684\u7522\u54c1\n<\/p><\/blockquote>\n<h2 id=\"toc_7\">\u5f9e\u7db2\u9801\u5b58\u53d6\u8a18\u9304\u77ad\u89e3\u4f7f\u7528\u8005\u884c\u70ba\u8207\u7db2\u9801\u5340\u584a\u8ca2\u737b\u5206\u6790-\u5be6\u6230\u7bc7<\/h2>\n<ul>\n<li>\u9801\u9762\u677f\u584a\u8ca2\u737b\u5206\u6790\u65b9\u6cd5<\/li>\n<li>\u8655\u7406\u8207\u5132\u5b58 web access log<\/li>\n<li>Ref \u8cc7\u8a0a\n<ul>\n<li>\u5728 url \u52a0\u4e0a ?ref=from<em>which<\/em>page<em>block \u8cc7\u8a0a (\u9801\u9762\u4ee3\u78bc<\/em>\u5340\u584a\u4ee3\u78bc_\u9806\u5e8f)<\/li>\n<li>\u8003\u984c: \u5206\u6790 ref \u8cc7\u6599<\/li>\n<li>\u6539\u9032: ?ref=\u4e3b\u984c\u9928<em>\u7db2\u9801\u4ee3\u78bc<\/em>KEY<em>\u5340\u584a\u540d\u7a31<\/em>\u9806\u5e8f_\u677f\u584a\u4f4d\u7f6e\u4ee3\u78bc<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<blockquote><p>\n\u70ba\u4f55\u4e0d\u7528 heap map \u6709\u73fe\u6210\u7684 3rd party \u5de5\u5177? \u6bd4\u8f03\u4e0d\u6e96? \u4f46\u99ac\u4e0a\u5c31\u6709\u4e86 XD\n<\/p><\/blockquote>\n<ul>\n<li>\u6539\u9032: \u52a0\u4e0a a\/b testing<\/li>\n<li>url ref \u554f\u984c: SEO \u592a\u7cdf\u7cd5\uff0c\u6539\u653e\u5728 data-rel \u88e1\u9762\uff0c\u7528 javascript \u53bb\u88dc rel \u5230 href \u5f8c\u9762\u53bb (workaroud)<\/li>\n<li>web access log \u6bd4 GA \u597d\u7684\u5730\u65b9: \u9801\u9762\u677f\u584a\u8ca2\u737b\u5206\u6790<\/li>\n<li>parse \u51fa\u7d00\u9304 timestamp, cookie<em>id, product<\/em>id, ref, page<em>log<\/em>id<\/li>\n<li>web servers -&gt; kafka cluster -&gt; DB(impala, postgreSQL \u6216 mysql)<\/li>\n<li>\u6574\u7406\u51fa\u5546\u54c1\u8f49\u63db\u8a08\u7b97 A-&gt;B (\u5f9e A \u5546\u54c1\u9801 -&gt; B)<\/li>\n<li>\u4f7f\u7528\u8005\u641c\u5c0b\u8207\u5546\u54c1\u9ede\u64ca (\u53ef\u4ee5\u5f37\u5316\u5168\u6587\u6aa2\u7d22)<\/li>\n<li>\u7d71\u8a08\u4f7f\u7528\u8005\u9ede\u64ca\u4f4d\u7f6e(\u4f8b\u5982\u662f\u6a19\u984cor\u5167\u6587or\u5716\u7247)<\/li>\n<li>ref \u597d\u8655: \u6e1b\u5c11\u4e0d\u5fc5\u8981\u5728 application \u88e1\u9762\u53bb\u63d2 code<\/li>\n<\/ul>\n<blockquote><p>\n\u9019\u662f\u4e00\u5834 pro tips \u7684 talk\n<\/p><\/blockquote>\n<ul>\n<li>RWD \u662f\u738b\u9053\uff0c\u4f46\u662f javascript \u8981\u642d\u914d media-query \u7d00\u9304\u7576\u6642\u7684\u7248\u578b\u662f\u54ea\u4e00\u7a2e\uff0c\u52a0\u5230 ref \u4e0a\u3002<\/li>\n<li>\u641c\u5c0b\u627e\u4e0d\u5230\u4ec0\u9ebc\u7684\u8cc7\u8a0a\u6709\u6642\u5019\u6bd4\u641c\u5c0b\u627e\u5230\uff0c\u66f4\u91cd\u8981\u3002\u8981\u600e\u9ebc\u627e\u51fa\u9019\u500b\u8cc7\u8a0a\u5462?<\/li>\n<\/ul>\n<h2 id=\"toc_8\">\u4f7f\u7528 Elasticsearch \u53ca Kibana \u9032\u884c\u5de8\u91cf\u8cc7\u6599\u641c\u5c0b\u53ca\u8996\u89ba\u5316<\/h2>\n<ul>\n<li>gogolook \u7684 Data Engineer<\/li>\n<li>why? \u5982\u4f55\u5c0d\u4ed8\u8001\u95c6\u4e00\u76f4 request \u8cc7\u6599\u554f\u984c? \u4eca\u5929 DAU \u591a\u5c11? \u5206\u570b\u5bb6? \u5206APP\u7248\u672c? \u770b\u4e00\u5e74? \u756b\u5716? \u7d66 raw data? \u4f7f\u7528\u8005\u665a\u7761\u55ce? \u4f7f\u7528\u8005\u6027\u5565?&#8230;etc<\/li>\n<\/ul>\n<blockquote><p>\n\u8ddf\u4e0a\u6b21\u770b\u5230\u7684 500px \u6848\u4f8b\u4e00\u6a23\uff0c\u5efa\u69cb\u4e00\u500b \u8cc7\u6599\u5206\u6790\u67e5\u8a62 \u7684\u5167\u90e8\u7cfb\u7d71\u3002\u800c Data Engineer \u8ca0\u8cac\u505a ETL -&gt; \u67e5\u8a62\u5e73\u53f0\u548c\u5716\u8868\n<\/p><\/blockquote>\n<ul>\n<li>\u4f7f\u7528 Kibana<\/li>\n<\/ul>\n<blockquote><p>\nKibana \u611f\u89ba\u662f\u6cdb\u7528\u578b\u7684\u4ecb\u9762\uff0c\u9084\u662f\u6709 20% \u7684\u554f\u984c\u89e3\u4e0d\u4e86\n<\/p><\/blockquote>\n<ul>\n<li>\u6709\u7528 Fluentd<\/li>\n<li>kibana \u7f3a\u9ede: No SQL JOIN 3. no subquery <\/li>\n<li>\u6240\u4ee5\u5269\u4e0b 20% \u7684\u554f\u984c\u5728\u8003\u616e\u7528 google bigquery \u6216 aws redshirt \u52a0\u4e0a tableau or periscope <\/li>\n<\/ul>\n<blockquote><p>\n\u6211\u89ba\u5f97\u6b63\u89e3\u9084\u662f\u4e56\u4e56\u4f5c data warehouse + \u8996\u89ba\u5316\n<\/p><\/blockquote>\n<h2 id=\"toc_9\">Visualization over Web<\/h2>\n<p><iframe loading=\"lazy\" title=\"Visualization over Web: Tools and Tips\uff0d\u5433\u6cf0\u8f1d\" src=\"https:\/\/www.slideshare.net\/slideshow\/embed_code\/key\/tv7o6g8yRcP7sh\" width=\"427\" height=\"356\" frameborder=\"0\" marginwidth=\"0\" marginheight=\"0\" scrolling=\"no\" style=\"border:1px solid #CCC; border-width:1px; margin-bottom:5px; max-width: 100%;\" allowfullscreen> <\/iframe> <\/p>\n<div style=\"margin-bottom:5px\"> <strong> <a href=\"https:\/\/www.slideshare.net\/tw_dsconf\/visualization-over-web-tools-and-tips\" title=\"Visualization over Web: Tools and Tips\uff0d\u5433\u6cf0\u8f1d\" target=\"_blank\">Visualization over Web: Tools and Tips\uff0d\u5433\u6cf0\u8f1d<\/a> <\/strong> from <strong><a href=\"https:\/\/www.slideshare.net\/tw_dsconf\" target=\"_blank\">\u53f0\u7063\u8cc7\u6599\u79d1\u5b78\u5e74\u6703<\/a><\/strong> <\/div>\n<ul>\n<li>D3.js<\/li>\n<li>why web<\/li>\n<li>responsive design<\/li>\n<\/ul>\n<blockquote><p>\n\u55ae\u8b1b\u5de5\u5177\u679c\u7136\u6bd4\u8f03\u7121\u804a\uff0c\u5f8c\u4f86\u5c31\u7761\u8457\u4e86&#8230;\n<\/p><\/blockquote>\n<h2 id=\"toc_10\">\u4ea4\u901a\u7db2\u8def\u89e3\u6790<\/h2>\n<ul>\n<li>\u8d85\u5f37\uff0c\u5305\u542b\u5be6\u52d9\u3001\u6a21\u578b\uff0c\u5305\u54e1\u5982\u4f55\u89e3\u6c7a\u8cc7\u6599\u4f86\u6e90\u554f\u984c\uff0c\u4f8b\u5982\u516c\u8eca\u53ea\u6709\u4e0a\u8eca\u8cc7\u6599\uff0c\u6c92\u6709\u4e0b\u8eca<\/li>\n<li>90% \u7684\u5237\u5361\u4ea4\u6613 = 90% \u7684\u62bd\u6a23\u6bd4\u4f8b<\/li>\n<\/ul>\n<blockquote><p>\n\u505a\u5927\u6578\u64da\uff0c\u624b\u4e0a\u63e1\u6709\u8cc7\u6599\u624d\u662f\u95dc\u9375\u554a\u554a\n<\/p><\/blockquote>\n<h2 id=\"toc_11\">\u8edf\u5de5\u4eba\u7684\u8cc7\u6599\u79d1\u5b78\u5947\u822a\uff0d\u7dda\u4e0a\u904a\u6232\u3001\u7db2\u8def\u5b78\u7fd2\u8207\u4e2d\u83ef\u8077\u68d2<\/h2>\n<p><iframe loading=\"lazy\" title=\"\u8edf\u5de5\u4eba\u7684\u8cc7\u6599\u79d1\u5b78\u5947\u822a\uff0d\u7dda\u4e0a\u904a\u6232\u3001\u7db2\u8def\u5b78\u7fd2\u8207\u4e2d\u83ef\u8077\u68d2 by \u8a31\u61f7\u4e2d\" src=\"https:\/\/www.slideshare.net\/slideshow\/embed_code\/key\/JYdjLArCxNuu5n\" width=\"427\" height=\"356\" frameborder=\"0\" marginwidth=\"0\" marginheight=\"0\" scrolling=\"no\" style=\"border:1px solid #CCC; border-width:1px; margin-bottom:5px; max-width: 100%;\" allowfullscreen> <\/iframe> <\/p>\n<div style=\"margin-bottom:5px\"> <strong> <a href=\"https:\/\/www.slideshare.net\/tw_dsconf\/by-51965458\" title=\"\u8edf\u5de5\u4eba\u7684\u8cc7\u6599\u79d1\u5b78\u5947\u822a\uff0d\u7dda\u4e0a\u904a\u6232\u3001\u7db2\u8def\u5b78\u7fd2\u8207\u4e2d\u83ef\u8077\u68d2 by \u8a31\u61f7\u4e2d\" target=\"_blank\">\u8edf\u5de5\u4eba\u7684\u8cc7\u6599\u79d1\u5b78\u5947\u822a\uff0d\u7dda\u4e0a\u904a\u6232\u3001\u7db2\u8def\u5b78\u7fd2\u8207\u4e2d\u83ef\u8077\u68d2 by \u8a31\u61f7\u4e2d<\/a> <\/strong> from <strong><a href=\"https:\/\/www.slideshare.net\/tw_dsconf\" target=\"_blank\">\u53f0\u7063\u8cc7\u6599\u79d1\u5b78\u5e74\u6703<\/a><\/strong> <\/div>\n<ul>\n<li>\u4ea4\u5927 14 \u5e74: OOAD, \u8edf\u9ad4\u5de5\u7a0b<\/li>\n<li>\u5728\u9673\u6607\u7def\u8001\u5e2b\u7684\u5be6\u9a57\u5ba4<\/li>\n<li>\u8edf\u9ad4\u5de5\u7a0b \u8ddf \u8cc7\u6599\u79d1\u5b78&#8230;.. \u55ef\u6c92\u4ec0\u9ebc\u95dc\u4fc2<\/li>\n<li>\u7cbe\u901a\u8cc7\u6599\u5206\u6790\u7684\u7d93\u9a57: \u904a\u6232\u6a58\u5b50, \u5747\u4e00\u6559\u80b2\u5e73\u53f0,CPBL<\/li>\n<li>Q: \u4e00\u6b3e\u904a\u6232\u53ef\u4ee5\u6d3b\u591a\u4e45? (\u4e00\u822c\u53ea\u67094~9\u500b\u6708)\n<ul>\n<li>\u7531\u6578\u64da\u7b97\u51fa\u6c88\u8ff7\u5ea6 + \u5075\u6e2c\u81c9\u90e8\u8868\u60c5 =&gt; \u9810\u6e2c\u4e0b\u4e00\u6b3e\u904a\u6232\u6c88\u8ff7\u5ea6<\/li>\n<li>\u5f8c\u4f86\u904a\u6232\u6a58\u5b50\u8f49\u4f5c\u624b\u904a\uff0c\u6c92\u6d3e\u4e0a\u7528\u5834<\/li>\n<li>\u8868\u60c5: \u53ad\u60e1\u8868\u60c5\u6703\u6bd4\u8f03\u5f37\u8a0a\u865f<\/li>\n<\/ul>\n<\/li>\n<li>Q: \u5747\u4e00\u6559\u80b2\u5e73\u53f0 for K-15 \u7ffb\u8f49\u6559\u80b2\n<ul>\n<li>\u514d\u8cbb\u5e6b\u4ed6\u5011\u5206\u6790 XD<\/li>\n<li>\u80cc\u666f: \u5b78\u751f\u7a0b\u5ea6\u5dee\u5f88\u591a, \u9ecf\u8457\u5ea6\u4f4e<\/li>\n<li>\u554f\u984c: \u9700\u8981\u4e00\u500b\u5feb\u901f\u6b63\u78ba\u6709\u8da3\u53c8\u4e0d\u5f71\u97ff\u5b78\u7fd2\u610f\u9858\u7684\u8a55\u91cf\u65b9\u6cd5<\/li>\n<li>\u89e3\u6cd5: \u9069\u6027\u6e2c\u9a57\u3002\u7a0b\u5ea6\u5dee\u5f88\u591a\uff0c\u9700\u8981\u5f88\u5feb\u5224\u65b7\u7a0b\u5ea6\u7d66\u4e88\u9069\u7576\u7684\u7df4\u7fd2\u984c\uff0c\u800c\u4e0d\u6703\u5687\u5230\u4ed6<\/li>\n<li>\u7531\u7df4\u7fd2\u984c\u7d00\u9304(\u7b54\u5c0d\u7b54\u932f\u6b21\u6578)\u5efa\u7acb\u4f7f\u7528\u8005\u7df4\u7fd2\u984c\u6a21\u578b -&gt; \u7df4\u7fd2\u984c\u63a8\u85a6\u3002\u7d50\u679c\u53ea\u63d0\u5347 5% &#8230;orz<\/li>\n<li>\u9700\u8981\u66f4\u591a\u8cc7\u6599\u4f86\u5206\u6790:\n<ul>\n<li>\u6539\u7528\u77e5\u8b58\u5730\u5716\u5ea7\u6a19\u3001\u9806\u5e8f + \u7fa4\u773e\u5916\u5305\u53bb \u6bd4\u8f03\u7df4\u7fd2\u984c\u7684\u76f8\u4f3c\u5ea6, \u96e3\u5ea6\uff0c\u589e\u52a0\u66f4\u591a\u8cc7\u6599&#8230;etc<\/li>\n<li>Bounty Workers <\/li>\n<\/ul>\n<\/li>\n<li>\u7528 Random Forest \u53d6\u5f97 features \u7684\u91cd\u8981\u6027\n<ul>\n<li>\u9020\u51fa\u4e00\u500b tree \u662f\u6240\u6709\u6e2c\u9a57\u984c\u7684\u76f8\u4f3c\u5ea6\u548c\u96e3\u5ea6\u9806\u5e8f\u95dc\u4fc2<\/li>\n<li>\u6700\u91cd\u8981\u7684 feature \u9084\u662f\u77e5\u8b58\u5730\u5716&#8230;&#8230;. XDXD<\/li>\n<\/ul>\n<\/li>\n<li>\u76ee\u6a19:\n<ul>\n<li>\u7528\u5c11\u91cf\u7684\u984c\u76ee\uff0c\u5c31\u53ef\u4ee5\u77e5\u9053\u5b78\u751f\u7a0b\u5ea6<\/li>\n<li>\u63d0\u4f9b\u8001\u5e2b\u8996\u89ba\u5316\u8a55\u91cf\u5716\u8868<\/li>\n<li>\u9069\u7576\u8a55\u4f30\u5c31\u53ef\u4ee5\u63a8\u85a6\u7df4\u7fd2\u984c<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<blockquote><p>\n\u5982\u679c\u6e2c\u9a57\u984c\u6c92\u6709\u9019\u9ebc\u591a\u7684\u8a71\uff0c\u662f\u4e0d\u662f ML \u5176\u5be6\u6d3e\u4e0d\u4e0a\u7528\u5834???? \u9760\u4eba\u5de5\u5206\u5c31\u597d\u4e86\n<\/p><\/blockquote>\n<ul>\n<li>Q: \u4e2d\u83ef\u8077\u68d2\n<ul>\n<li>\u6230\u7e3e\u7403\u8ff7\u73fe\u8c61! \u627e\u51fa\u5f71\u97ff\u7968\u623f\u7684\u8981\u7d20<\/li>\n<li>\u8cc7\u6599\u5206\u6790\u6d41\u7a0b\n<ul>\n<li>\u8655\u7406\u8cc7\u6599\u6703\u82b1\u638970~80%\u6642\u9593<\/li>\n<li>\u8981\u5148\u6709\u554f\u984c\uff0c\u624d\u6709\u8cc7\u6599\u79d1\u5b78<\/li>\n<li>\u95dc\u9375: \u8981\u6709\u300c\u60f3\u50cf\u529b\u300d\uff0c\u8981\u6709\u5f88\u591a\u554f\u984c\u3002\u4e0d\u65b7\u554f\u554f\u984c\uff0c\u5206\u6790\u8cc7\u6599\uff0c\u7136\u5f8c\u8a6e\u91cb\u7d50\u679c<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<\/li>\n<li>\u8cc7\u6599\u79d1\u5b78 != big data\u3002\u5747\u4e00\u4e5f\u624dG<\/li>\n<li>\u8cc7\u6599\u5929\u751f\u6709\u7f3a\u9677<\/p>\n<\/li>\n<li>\n<p>QA: \u8ddf\u5929\u6c23\u3001\u98a8\u5411\u4e5f\u6709\u95dc\u4fc2    <\/p>\n<\/li>\n<\/ul>\n<blockquote>\n<p>\u9019\u5834\u883b\u7cbe\u5f69\u6709\u8da3\u7684\uff0c\u96d6\u7136\u958b\u5834\u7684\u8edf\u5de5\u771f\u7684\u6c92\u4ec0\u9ebc\u95dc\u4fc2<br \/>\n\u8edf\u9ad4\u5de5\u7a0b\u8ddf\u8cc7\u6599\u79d1\u5b78\u7684\u95dc\u4fc2\u662f&#8230;&#8230; \u6c92\u4ec0\u9ebc\u95dc\u4fc2\n<\/p><\/blockquote>\n<h2 id=\"toc_12\">\u5176\u4ed6<\/h2>\n<ul>\n<li>Party \u73fe\u5834\u97f3\u6a02\u6709\u9ede\u6d6a\u8cbb<\/li>\n<li>\u8d08\u54c1\u676f\u5b50\u4e0d\u932f<\/li>\n<li>\u9910\u9ede\u90fd\u4e0d\u932f<\/li>\n<li>\u6574\u9ad4\u8a55\u50f9\u9084\u4e0d\u932f\uff0c\u660e\u5e74\u9084\u662f\u53ef\u4ee5\u4f86<\/li>\n<li>\u4e0d\u904e\u6c92\u6709\u53bb\u5e74\u611f\u52d5\u5c31\u662f\u4e86\uff0c\u4eca\u5e74\u7684\u7684 talk \u7b2c\u4e00\u5929\u592a\u591a\u6559\u6388\u4e86<\/li>\n<li>\u5e0c\u671b\u591a\u4e00\u9ede\u696d\u754c\u7d93\u9a57\u5206\u4eab\uff0c\u4f8b\u5982\u53bb\u5e74\u7684 gogolook \u5c31\u8d85\u7d1a\u8d0a<\/li>\n<li>\u8ddf\u53bb\u5e74\u4e00\u6a23\uff0c\u6295\u5f71\u7247\u99ac\u4e0a\u5c31\u653e\u51fa\u4f86! \u8d0a<\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>\u4eca\u5e74 2015\/8\/22-23 \u53bb\u53c3\u52a0\u4e86\u53f0\u7063\u8cc7\u6599\u79d1\u5b78\u611b\u597d\u8005\u5e74\u6703\uff0c\u9019\u662f\u7576\u6642\u7684\u7b46\u8a18\u6574\u7406\u3002\u53e6\u5916\u9084\u6709 \u6703\u773e\u7248\u7684 hac &hellip; <\/p>\n<p class=\"link-more\"><a href=\"https:\/\/ihower.tw\/blog\/8440-%e5%8f%b0%e7%81%a3%e8%b3%87%e6%96%99%e7%a7%91%e5%ad%b8%e6%84%9b%e5%a5%bd%e8%80%85%e5%b9%b4%e6%9c%83-2015-%e5%bf%83%e5%be%97%e7%ad%86%e8%a8%98\" class=\"more-link\">\u95b1\u8b80\u5168\u6587<span class=\"screen-reader-text\">\u3008\u53f0\u7063\u8cc7\u6599\u79d1\u5b78\u611b\u597d\u8005\u5e74\u6703 2015 \u5fc3\u5f97\u7b46\u8a18\u3009<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"jetpack_post_was_ever_published":false,"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2}},"categories":[74],"tags":[],"class_list":["post-8440","post","type-post","status-publish","format-standard","hentry","category-data-science","entry"],"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_shortlink":"https:\/\/wp.me\/p1q6tG-2c8","jetpack_sharing_enabled":true,"jetpack_likes_enabled":true,"_links":{"self":[{"href":"https:\/\/ihower.tw\/blog\/wp-json\/wp\/v2\/posts\/8440","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/ihower.tw\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/ihower.tw\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/ihower.tw\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/ihower.tw\/blog\/wp-json\/wp\/v2\/comments?post=8440"}],"version-history":[{"count":8,"href":"https:\/\/ihower.tw\/blog\/wp-json\/wp\/v2\/posts\/8440\/revisions"}],"predecessor-version":[{"id":8450,"href":"https:\/\/ihower.tw\/blog\/wp-json\/wp\/v2\/posts\/8440\/revisions\/8450"}],"wp:attachment":[{"href":"https:\/\/ihower.tw\/blog\/wp-json\/wp\/v2\/media?parent=8440"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/ihower.tw\/blog\/wp-json\/wp\/v2\/categories?post=8440"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/ihower.tw\/blog\/wp-json\/wp\/v2\/tags?post=8440"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}