tag:blogger.com,1999:blog-41968416487007777182024-03-19T02:29:23.826-07:00Maximize Productivity with Industrial Engineer and Operations Research Toolslarrydaghttp://www.blogger.com/profile/06592231655463233562noreply@blogger.comBlogger204125tag:blogger.com,1999:blog-4196841648700777718.post-1987562805631656132017-03-27T20:28:00.000-07:002017-03-27T20:28:33.260-07:00Introducing Boardgame Recommendation website built with ShinyMy family and I love boardgames. We are also on the lookout for new ones that would fit with the ones we already like to play. So I decided to create a boardgame recommendation website<br />
<br />
<a href="https://larrydag.shinyapps.io/boardgame_reco/">https://larrydag.shinyapps.io/boardgame_reco/</a><br />
<br />
I built it using R. The recommendation engine uses a very simple collaborative filtering algortihm based on correlation scores from other boardgame players collection lists. The collections are gathered using the API from <a href="http://boardgamegeek.com/">BoardgameGeek.com</a>. It is very much in a beta project phase as I just wanted to get something built to get working. <br />
<br />
I also wanted another project to build in <a href="http://shiny.rstudio.com/">Shiny</a>. I really like how easy it is to publish R projects with Shiny. <br />
<br />
Some of the features include:<br />
<br />
<ul>
<li>Ability to enter your own collection</li>
<li>Get recommendation on your collection</li>
<li>Amazon link to buy boardgame that is recommended</li>
</ul>
<div>
Its a work in progress. There is much to clean up and to make more presentable. Please take a look and offer comments to help improve the website. </div>
larrydaghttp://www.blogger.com/profile/06592231655463233562noreply@blogger.com163tag:blogger.com,1999:blog-4196841648700777718.post-81350554049141298032016-11-16T10:55:00.000-08:002016-11-16T10:55:11.985-08:00Microsoft allows trying its new SQL Server for LinuxThis is a great time to have tools to do data analysis. Microsoft is now <a href="https://www.microsoft.com/en-us/evalcenter/evaluate-sql-server-vnext-ctp">allowing evaluations</a> of its new SQL Server. This new SQL Server can now be deployed to a Linux OS environment. This is big news especially for those that have followed this blog through the years.<br />
<br />
The new SQL Server version is also carries new analytic tools such as R. From this Techcrunch <a href="https://techcrunch.com/2016/11/16/microsofts-sql-server-for-linux-is-now-available-for-testing/">announcement</a><br />
<br />
<blockquote class="tr_bq">
<span style="background-color: #f9f9f9; color: #3e433e; font-family: "Open Sans", HelveticaNeue, Helvetica, Arial, sans-serif; font-size: 16px; line-height: 26px;">The new version of SQL Server will include improved support for R Services and a number of new machine learning and deep neural networking features.</span></blockquote>
<br />
These are exciting times in deed. I hope Microsoft continues this path to bringing new tools to the market.larrydaghttp://www.blogger.com/profile/06592231655463233562noreply@blogger.com42tag:blogger.com,1999:blog-4196841648700777718.post-49494360294519410392016-10-17T07:04:00.001-07:002016-10-17T07:04:37.083-07:00Microsoft releases LightGBMMicrosoft has been really increasing their development of tools in the predictive analytics and machine learning space. Another such tool they released recently is <a href="https://github.com/Microsoft/LightGBM">LightGBM</a>. From the Github site...<br />
<br />
<blockquote class="tr_bq">
<span style="background-color: white; color: #666666; font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol"; font-size: 16px; line-height: 24px;">LightGBM is a fast, distributed, high performance gradient boosting (GBDT, GBRT, GBM or MART) framework based on decision tree algorithms, used for ranking, classification and many other machine learning tasks.</span></blockquote>
<br />
Microsoft is definitely increasing their attempts to capitalize on the machine learning and big data movement. I hope they continue to develop tools such as LightGBM and <a href="https://msdn.microsoft.com/en-us/library/mt604845.aspx">R with SQL Server</a>. larrydaghttp://www.blogger.com/profile/06592231655463233562noreply@blogger.com128tag:blogger.com,1999:blog-4196841648700777718.post-84569433257592403892016-01-13T05:43:00.001-08:002016-01-13T05:43:10.862-08:00Revolution R is now Microsoft RThere has been a huge shift in the force. Can you feel it? Today Microsoft is announcing that Revolution R is now Microsoft R. This <a href="https://mran.revolutionanalytics.com/open/">includes the Open R version</a> as well. Some notable features of Microsoft Open R.<br />
<br />
<br />
<ul>
<li>Open R 3.2.2 is fully compatible with R 3.2.2</li>
<li>Microsoft, Mac OS X, and Linux support (wow!)</li>
<li>Available free to download</li>
<li>Multi-threaded math libraries</li>
<li>Enterprise version available (for a price)</li>
</ul>
<div>
<br /></div>
<div>
This is an interesting shift for Microsoft in the analytics space. Microsoft is flexing it's server muscles to show that it can play in the data science field as well. I'm wondering if the industry is going to shift. We know that <a href="http://www.channelpartnersonline.com/news/2015/07/ibm-oracle-sas-named-leaders-in-growing-cloud-analytics-market.aspx">IBM, SAS, and Oracle</a> are pushing forward with cloud analytics. Microsoft is showing a major commitment to partner with the R community. I think this is a bold and wise move as R has shown nothing but growth the last few years.</div>
larrydaghttp://www.blogger.com/profile/06592231655463233562noreply@blogger.com39tag:blogger.com,1999:blog-4196841648700777718.post-26172999651424376292014-12-19T19:05:00.002-08:002014-12-19T19:06:03.305-08:00Statistical Analysis and Data Mining hot on LinkedIn for 2014Statistical Analysis and Data Mining are considered the hottest skills on LinkedIn for 2014. According to <a href="http://talent.linkedin.com/blog/index.php/2014/12/the-25-hottest-skills-to-recruit-for-on-linkedin">their report</a> from analyzing jobs and recruiters on the LinkedIn website. I would say its safe to say that it will continue to be hot for 2015.<br />
<br />
If you are looking to get your skills honed up I would suggest looking at ComputerWorld's <a href="http://www.computerworld.com/article/2497143/business-intelligence-beginner-s-guide-to-r-introduction.html">Beginners Guide to R</a>. It looks like a complete tutorial and is indexed rather well.larrydaghttp://www.blogger.com/profile/06592231655463233562noreply@blogger.com22tag:blogger.com,1999:blog-4196841648700777718.post-30394214549134208332014-09-22T18:25:00.003-07:002014-09-22T18:26:14.349-07:00Interesting high contrast plots in RI was inspired by <a href="http://nxn.se/post/97650612370/high-contrast-stacked-distribution-plots">this blog post</a> and thought I could do the same thing in R. Well I posted the code in<a href="https://plus.google.com/100136476859853724604/posts/VHCfVnn1aPH"> Google+</a><br />
<br />
<br />
Here are my results. Not bad.<br />
<br />
<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiwbntsuq_A2GBE2VlEHMF5vKeZH6juhpM37hY8OMer0Tl9I9fT51su1CfOO7Uzz8ty4COkfbz7NIAlAYrpVFlrPu8LeSvPKytI7IvxbL3Z1LqXyCFast61EyCWQJ2lDnjidPLQUZgnFF4/s1600/Rplot.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiwbntsuq_A2GBE2VlEHMF5vKeZH6juhpM37hY8OMer0Tl9I9fT51su1CfOO7Uzz8ty4COkfbz7NIAlAYrpVFlrPu8LeSvPKytI7IvxbL3Z1LqXyCFast61EyCWQJ2lDnjidPLQUZgnFF4/s1600/Rplot.png" height="363" width="400" /></a></div>
<br />larrydaghttp://www.blogger.com/profile/06592231655463233562noreply@blogger.com11tag:blogger.com,1999:blog-4196841648700777718.post-51381844460243277822014-08-04T18:09:00.001-07:002014-08-04T18:13:20.173-07:00Introducing the Shiny App DThiringWell it has a been a long time since I have written anything on this blog. I am long overdue. I've been terribly busy learning new things and getting on with life. One of the things I have learned is building R applications using Shiny developed by <a href="http://www.rstudio.com/">RStudio</a>. The folks at RStudio have also created a way to deploy Shiny apps using Shinyapps.io. Follow the link to DThiring<br />
<br />
<a href="http://larrydag.shinyapps.io/dthiring/">http://larrydag.shinyapps.io/dthiring/</a><br />
<br />
I'm a big fan of Data Tau which is a Data Science equivalent to Hacker News. Like Hacker News, Data Tau has a Who's Hiring comment on the first of every month. It is a good resource for those looking to see relevant jobs in the Data Science world. Well someone created a hiring listings aggregator called <a href="http://hnhiring.me/">http://hnhiring.me/</a>. I decided I wanted to see if I could build a similar application using R and Shiny. <br />
<br />
I will be posting the source code for this app to github in the near future.<br />
<br />
If you have any ideas on how to improve this application let me know in this comment section. larrydaghttp://www.blogger.com/profile/06592231655463233562noreply@blogger.com62tag:blogger.com,1999:blog-4196841648700777718.post-29883632526648319992014-03-15T05:40:00.003-07:002014-03-15T05:40:36.523-07:00OpenOpt 0.53<br />
I'm glad to inform you about new OpenOpt Suite release 0.53:<br /> <a href="http://openopt.org/StochasticProgramming">Stochastic programming addon</a> now is available for free<br /> Some minor changes<br />
Regards, Dmitrey.Dmitreyhttp://www.blogger.com/profile/18259817544023299492noreply@blogger.com5tag:blogger.com,1999:blog-4196841648700777718.post-87873996135612584012013-12-15T04:45:00.001-08:002013-12-15T04:45:15.432-08:00OpenOpt Suite release 0.52I'm glad to inform you about new OpenOpt Suite release 0.52 (2013-Dec-15):<br />
Minor <a href="http://openopt.org/interalg">interalg</a> speedup<br /> oofun expression<br /> MATLAB solvers <a href="http://openopt.org/fmincon">fmincon</a> and fsolve have been connected<br /> Several MATLAB <a href="http://openopt.org/ODE">ODE</a> solvers have been connected<br /> New ODE solvers, parameters abstol and reltol<br /> New <a href="http://openopt.org/GLP">GLP</a> solver: direct<br /> Some minor bugfixes and improvements <br />
Regards, Dmitrey.Dmitreyhttp://www.blogger.com/profile/18259817544023299492noreply@blogger.com4tag:blogger.com,1999:blog-4196841648700777718.post-32302083265789257552013-09-15T07:51:00.002-07:002013-09-15T07:51:28.274-07:00New OpenOpt Suite release 0.51<span class="xfm_3536600425"><span class="xfmc0"></span></span><br />
New OpenOpt suite v 0.51 has been released:<br />
<ul>
<li> Some improvements for <a href="http://openopt.org/FuncDesigner" target="_blank">FuncDesigner</a> automatic differentiation and <a href="http://openopt.org/QP" title="QP">QP</a>
</li>
<li> FuncDesigner now can model sparse (MI)(QC)QP
</li>
<li> Octave <a href="http://openopt.org/QP" title="QP">QP</a> solver has been connected
</li>
<li> <a href="http://openopt.org/MATLAB" title="MATLAB">MATLAB</a> solvers linprog (<a href="http://openopt.org/LP" title="LP">LP</a>), quadprog (<a href="http://openopt.org/QP" title="QP">QP</a>), lsqlin (<a href="http://openopt.org/LLSP" title="LLSP">LLSP</a>), bintprog (<a href="http://openopt.org/MILP" title="MILP">MILP</a>)
</li>
<li> New <a href="http://openopt.org/NLP" title="NLP">NLP</a> solver: <a href="http://openopt.org/knitro" title="knitro">knitro</a>
</li>
<li> Some elements of 2nd order interval analysis, mostly for interalg
</li>
<li> Some <a href="http://openopt.org/interalg" title="interalg">interalg</a> improvements
</li>
<li> <a href="http://openopt.org/interalg" title="interalg">interalg</a> can directly handle (MI)LP and (possibly nonconvex) (MI)(QC)QP
</li>
<li> New classes: knapsack problem (<a href="http://openopt.org/KSP" title="KSP">KSP</a>), bin packing problem (<a href="http://openopt.org/BPP" title="BPP">BPP</a>), dominating set problem (<a href="http://openopt.org/DSP" title="DSP">DSP</a>)
</li>
<li> FuncDesigner can model <a href="http://openopt.org/SOCP" title="SOCP">SOCP</a>
</li>
<li> <a href="http://openopt.org/SpaceFuncs" target="_blank">SpaceFuncs</a> has been adjusted for recent versions of Python and <a href="http://openopt.org/NumPy" title="NumPy">NumPy</a></li>
</ul>
visit <a href="http://openopt.org/" target="_blank">http://openopt.org</a> for more details.
<img alt="" src="https://secureurl.un.net.ua/url/tv6FTuhobz5S9dxWgNWkug/aHR0cDovL2ZyZWVtYWlsLnVrci5uZXQvYXBpL3B1YmxpYy9tZXNzYWdlX3JlYWQ_YT1nS3VpN0lobktIbXNrN3V2WkxFcFdTVHRDTmM4RTh6bFUxTkxjcm5HSVJTdWZrY1VuSldlOUs1am43dURZbWpqdzM5bWE3V2QtS3ZSdy1Sb1p0MUdjTmNjWkRPdGtQVHkteHBGSDBNPQ?orig_uri=http://freemail.ukr.net/api/public/message_read?a=gKui7IhnKHmsk7uvZLEpWSTtCNc8E8zlU1NLcrnGIRSufkcUnJWe9K5jn7uDYmjjw39ma7Wd-KvRw-RoZt1GcNccZDOtkPTy-xpFH0M=" />Dmitreyhttp://www.blogger.com/profile/18259817544023299492noreply@blogger.com13tag:blogger.com,1999:blog-4196841648700777718.post-90298480472527658272013-06-15T08:07:00.004-07:002013-06-15T08:07:54.732-07:00new OpenOpt Suite release 0.50<div class="entry-content">
Hi all,<br />
I'm glad to inform you about new OpenOpt Suite release 0.50 (2013-June-15):<br />
<br />
* <a href="http://openopt.org/interalg">interalg</a>
(solver with specifiable accuracy) now works many times (sometimes
orders) faster on (possibly multidimensional) integration problems (<a href="http://openopt.org/">IP</a>) and on some optimization problems<br /> * Add modeling dense (MI)(QC)<a href="http://openopt.org/QP">QP</a> in FuncDesigner (alpha-version, rendering may work slowly yet)<br /> * Bugfix for <a href="http://openopt.org/cplex">cplex</a> wrapper<br /> * Some improvements for FuncDesigner interval analysis (and thus interalg)<br /> * Add FuncDesigner interval analysis for <em>tan</em> in range(-pi/2,pi/2)<br /> * Some other bugfixes and improvements<br /> * (Proprietary) <a href="http://openopt.org/StochasticProgramming">FuncDesigner stochastic addon</a> now is available as standalone <em>pyc</em>-file, became available for Python3 as well <br />
<br />
Regards, Dmitrey.<br />
</div>
<br />Dmitreyhttp://www.blogger.com/profile/18259817544023299492noreply@blogger.com4tag:blogger.com,1999:blog-4196841648700777718.post-43918939715154277172013-03-15T06:25:00.001-07:002013-03-15T06:25:53.661-07:00OpenOpt Suite release 0.45I'm glad to inform you about new OpenOpt Suite release 0.45 (2013-March-15):<br /> * Essential improvements for <a href="http://openopt.org/FuncDesigner" target="_blank">FuncDesigner</a> interval analysis (thus affect <a href="http://openopt.org/interalg" target="_blank">interalg</a>)<br />
* Temporary walkaround for a serious bug in FuncDesigner automatic
differentiation kernel due to a bug in some versions of Python or NumPy,
may affect optimization problems, including (MI)LP, (MI)NLP, TSP etc<br /> * Some other minor bugfixes and improvements <br />
<br />Dmitreyhttp://www.blogger.com/profile/18259817544023299492noreply@blogger.com23tag:blogger.com,1999:blog-4196841648700777718.post-24342655893698481122013-02-16T07:32:00.000-08:002013-02-16T12:05:09.158-08:00Google Statistician uses R and other programming toolsA great interview on the Simply Statistics blog with <a href="http://simplystatistics.org/2013/02/15/interview-with-nick-chamandy-statistician-at-google/">Google's Nick Chamandy</a>, Phd in Statistics. Explains that he mainly uses R among other tools to perform his work at Google. Also of note is the active data science community within Google that uses R as well as some other interesting tools. Note that they use a lot of data at Google, understandably, and that R usually can not handle the size. They do a lot of ad hoc reduction of the data with tools like map reduce, Go, and even an R API. I would love to see how they use the R API to assimilate data.<br />
<br />
An interesting insight from the interview is the amount of programming done by the Statisticians. It seems the culture at Google is to foster autonomy and let the modelers develop their own data manipulation from the raw data. This requires a broader skillset beyond the statistical analysis tools.<br />
<br />
I've found in my work that having knowledge in many tools like R, CPLEX, and GLPK allows me to be a more effective in my work. Recently I've been learning a lot of SQL using the <a href="http://www.postgresql.org/">PostgreSQL</a> platform. The tools of SQL combined with statistical tools like R make for a very strong combination. I'm very agile in my work and can do a varied number of decision analysis.larrydaghttp://www.blogger.com/profile/06592231655463233562noreply@blogger.com5tag:blogger.com,1999:blog-4196841648700777718.post-63703502998757889692012-12-15T07:16:00.001-08:002012-12-15T07:16:25.826-08:00OpenOpt Suite release 0.43<span class="xfmb"><span>
</span></span><br />
<div class="xfmc0">
I'm glad to inform you about new OpenOpt release 0.43 (2012-Dec-15):<br />
<br /> * <a href="http://openopt.org/interalg">interalg</a> now can solve SNLE in 2nd mode (parameter dataHandling = "raw", before - only "sorted")<br /> * Many other improvements for interalg<br /> * Some improvements for <a href="http://funcdesigner/">FuncDesigner</a> kernel<br /> * FuncDesigner <a href="http://openopt.org/ODE">ODE</a>
now has 3 arguments instead of 4 (backward incompatibility!), e.g. {t:
np.linspace(0,1,100)} or mere np.linspace(0,1,100) if your ODE right
side is time-independend<br /> * FuncDesigner <a href="http://openopt.org/StochasticProgramming">stochastic addon</a> now can handle some problems with gradient-based <a href="http://openopt.org/NLP">NLP</a> / <a href="http://openopt.org/NSP">NSP</a> solvers<br /> * Many minor improvements and some bugfixes<br />
<br />
Visit <a href="http://openopt.org/" target="_blank">openopt.org</a> for more details.<br />
<br />
Regards, D.</div>
<br />
<pre>
</pre>
Dmitreyhttp://www.blogger.com/profile/18259817544023299492noreply@blogger.com0tag:blogger.com,1999:blog-4196841648700777718.post-83614972842622500942012-09-25T13:12:00.000-07:002012-09-27T09:51:16.541-07:00Day in the life of a Data ScientistA great read from the Decomposition blog about the <a href="http://seanjtaylor.com/2012/09/18/the-data-science-loop/">day in the life of a Data Scientist</a>. I consider <a href="http://industrialengineertools.blogspot.com/search?q=data+scientist">myself a Data Scientist</a> by any other name. The blog article by Sean does a great job of breaking down the essence of making better decisions for the organization you may be involved.<br />
<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhGU9J_fOaLWFbfb-XWK5OGRm_AQWiIsRCkgRHuySgqdRp57jKWY1MY2nwzqtFK-GhtEbh1bxuyPVm7WmYe69Ydbb-vzuc124R2L-qIE-CmA3lfJjWG6buu2Cioj3TRdRi_9GwHx5tjqD8/s1600/frontpagecolumbo1.jpg" imageanchor="1" style="clear: left; float: left; margin-bottom: 1em; margin-right: 1em;"><img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhGU9J_fOaLWFbfb-XWK5OGRm_AQWiIsRCkgRHuySgqdRp57jKWY1MY2nwzqtFK-GhtEbh1bxuyPVm7WmYe69Ydbb-vzuc124R2L-qIE-CmA3lfJjWG6buu2Cioj3TRdRi_9GwHx5tjqD8/s1600/frontpagecolumbo1.jpg" /></a></div>
<br />
I've always thought asking good questions is the start of good analysis. The organizations basically doesn't know what it doesn't know. A good Data Scientist will be a like a sleuth looking for clues. In all honesty that may be the most fun about being a Data Scientist.larrydaghttp://www.blogger.com/profile/06592231655463233562noreply@blogger.com2tag:blogger.com,1999:blog-4196841648700777718.post-62540825120603666692012-09-15T04:13:00.000-07:002012-09-15T04:13:14.523-07:00OpenOpt Suite release 0.42<span class="xfmb"><span>
</span></span><br />
<div class="xfmc0">
Hi all,<br />
<br />
I'm glad to inform you about new OpenOpt Suite release 0.42 (2012-Sept-15), fa free Python-written cross-platform software with primal focus on numerical optimization. Main changes:<br />
<br />
* Some improvements for solver <a href="http://openopt.org/interalg">interalg</a>, including handling of categorical variables<br /> * Some parameters for solver <a href="http://openopt.org/gsubg">gsubg</a><br /> * Speedup objective function for de and pswarm on <a href="http://openopt.org/FuncDesigner">FuncDesigner</a> models<br /> * New global (<a href="http://openopt.org/GLP">GLP</a>) solver: asa (adaptive simulated annealing)<br /> * Some new classes for network problems: <a href="http://openopt.org/TSP">TSP</a> (traveling salesman problem), <a href="http://openopt.org/STAB">STAB</a> (maximum graph stable set)], <a href="http://openopt.org/MCP">MCP</a> (maximum clique problem)<br /> * Improvements for FD XOR (and now it can handle many inputs)<br /> * Solver de has parameter "seed", also, now it works with <a href="http://openopt.org/PyPy">PyPy</a><br /> * Function <b>sign</b> now is available in FuncDesigner<br />
* FuncDesigner interval analysis (and thus solver interalg) now can handle non-monotone splines of 1st order<br /> * FuncDesigner now can handle parameter fixedVars as Python dict<br />
* Now scipy InterpolatedUnivariateSpline is used in FuncDesigner
interpolator() instead of UnivariateSpline. This creates backward
incompatibility - you cannot pass smoothing parameter (s) to
interpolator no longer.<br /> * <a href="http://openopt.org/SpaceFuncs">SpaceFuncs</a>: add Point weight, Disk, Ball and method contains(), bugfix for importing Sphere, some new examples<br /> * Some improvements (essential speedup, new parameter <b>interpolate</b> for <b>P()</b>) for our (currently commercial) FuncDesigner <a href="http://openopt.org/StochasticProgramming">Stochastic Programming</a> addon<br /> * Some bugfixes<br />
<br />
In our website ( <a href="http://openopt.org/">http://openopt.org</a> ) you could vote for most required OpenOpt Suite development direction(s) (poll has been renewed, previous results are <a href="http://www.easypolls.net/poll.html?p=4fca54a3e4b05091b02c35ad">here</a>).<br />
<br />
Regards, D. </div>
Dmitreyhttp://www.blogger.com/profile/18259817544023299492noreply@blogger.com3tag:blogger.com,1999:blog-4196841648700777718.post-81618706912305413392012-09-10T06:00:00.000-07:002012-09-10T06:00:05.072-07:00Upgrade your skill sets with free coursesWe are in the midst of the <a href="http://industrialengineertools.blogspot.com/2011/05/welcome-to-insight-age.html">Insight Age</a>. We have moved beyond capturing data and are now processing information. Properly processing the large amounts of data requires knowlege and skill sets. Fortunately there are many ways to develop those skills.<br />
<br />
<a href="http://www.class-central.com/">Class Central</a> is a website that provides a complete list of free online courses from some of the most established and prestigious universities in the world. Websites like these are helping to make the world smaller by providing free and accessible learning resources.<br />
<br />
<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgUrGlgBDkZUnmVR7LkYatb-EQob9qyCVMT9UlWFjGDdU5TSop6DfVUg0RNqv6A0ysyQY1XBdyt_3ldtsLYVpMRLrIztTJiEIBlpwaue62b8_vo6BqdtCNu-Cnulm3gjFyEMy1uU5kdLzo/s1600/class-central.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" height="187" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgUrGlgBDkZUnmVR7LkYatb-EQob9qyCVMT9UlWFjGDdU5TSop6DfVUg0RNqv6A0ysyQY1XBdyt_3ldtsLYVpMRLrIztTJiEIBlpwaue62b8_vo6BqdtCNu-Cnulm3gjFyEMy1uU5kdLzo/s400/class-central.png" width="400" /></a></div>
I am a big fan of <a href="http://industrialengineertools.blogspot.com/2010/04/operations-research-courses-via-open.html">open courseware</a>. There are plenty of other places to look for open coureses. The <a href="http://www.ocwconsortium.org/">Open Courseware Consortium</a> is a useful resource. A good metasearch site like <a href="http://www.opencontent.org/ocwfinder/">OpenCourseWare Finder</a> is valuable as well. larrydaghttp://www.blogger.com/profile/06592231655463233562noreply@blogger.com3tag:blogger.com,1999:blog-4196841648700777718.post-14077082784375824962012-07-02T06:08:00.000-07:002012-09-27T09:52:13.459-07:00Popularity of R continuesNo doubt those that read my blog know that the tools I use to do my Industrial Engineering and Operations Research work heavily rely on the open source side of software. That is why I try to support as many open source projects such as COIN-OR, GLPK, and <a href="http://industrialengineertools.blogspot.com/2012/06/openopt-suite-039.html">OpenOpt</a>. One tool that I love to perform Applied Math and Statistics is the <a href="http://www.r-project.org/">statistical computing platform R</a>. So it comes as no surprise that I like to see how R is growing and its popularity among programmers.<br />
<br />
A recent blog from RedMonk produced results of a <a href="http://redmonk.com/sogrady/2012/02/08/language-rankings-2-2012/">programming language popularity study</a>. The study involved ranking popularity using common social media online sites such as Stack Overflow and GitHub. These sites draw in a lot of programmers for their popularity around Q&A and code review. I was surprised to see that R ranks highly compared to some very prominant programming languages.<br />
<br />
<div class="separator" style="clear: both; text-align: center;">
<a href="http://img.skitch.com/20120208-jqctdbqg7yfbj785pe9gjtjp8t.medium.jpg" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" height="320" src="http://img.skitch.com/20120208-jqctdbqg7yfbj785pe9gjtjp8t.medium.jpg" width="400" /></a></div>
<br />
Also interesting to note that the only other "Data Science" type of programming language I could find was Matlab. As far as I could tell SAS, S, SPSS, Stata are still rather popular but apparently not among the programming community.larrydaghttp://www.blogger.com/profile/06592231655463233562noreply@blogger.com10tag:blogger.com,1999:blog-4196841648700777718.post-79533341442101735362012-06-15T03:56:00.002-07:002012-06-15T08:22:05.626-07:00OpenOpt Suite 0.39<div class="answer-body">
Hi all,<br />
<br />
I'm glad to inform you about new OpenOpt release 0.39 (quarterly since 2007).<br />
<br />
OpenOpt is free, even for commercial purposes, cross-platform
software for mathematical modeling and (mainstream) optimization. Our
website have reached 259 visitors daily, that is same to <a href="http://tomopt.com/">tomopt.com</a> and ~ 1/3 of <a href="http://gams.com/">gams.com</a> ( <a href="http://forum.openopt.org/viewtopic.php?id=570">details</a> ).<br />
<br />
In the new release:<br />
<ul>
<li><a href="http://openopt.org/interalg">interalg</a> (medium-scaled
solver with specifiable accuracy abs(f-f*) <= fTol): add categorical
variables and general logical constraints, many other improvements</li>
<li>Some improvements for automatic differentiation</li>
<li><a href="http://openopt.org/DerApproximator">DerApproximator</a> and some OpenOpt/FuncDesigner functionality now
works with PyPy (Python with dinamic compilation, some problems are
solved several times faster now)</li>
<li>New solver lsmr for dense/sparse <a href="http://openopt.org/LLSP">LLSP</a> (linear least squares)</li>
<li>Some bugfixes and some other changes </li>
</ul>
In our website (<a href="http://openopt.org/">openopt.org</a>) you could vote for most required OpenOpt Suite development direction(s).</div>Dmitreyhttp://www.blogger.com/profile/18259817544023299492noreply@blogger.com3tag:blogger.com,1999:blog-4196841648700777718.post-90939111568280202212012-05-21T06:50:00.000-07:002012-05-21T06:50:11.619-07:00National Registry of Exonerations charts with RAccording to recent news (<a href="http://www.dallasnews.com/news/community-news/dallas/headlines/20120520-new-national-database-analyzes-reasons-for-wrongful-convictions.ece?action=reregister">dallasnews.com</a>) there is a new release of a public national database for wrongful convictions. There are plenty of details in the public list including Age, Race, and how the conviction was overturned. According to the database it seems that most of the convictions were overturned due to DNA evidence.<br />
<br />
I thought it would be interesting to plot summaries of the details using the open source statistical computing environment R Project. The following are the plots from the National Registry of Exonerations database.<br />
<br />
<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgKdiOpX1eV-hd0mJwlnVQU0jcvtE8_0WaA35qw2JNFRpZ_H0lDVxB4jDXQqVup85YBdb7Yzb62kOa-jH4aU_8OgzOQfcGl_ZoGv1ygl1w1-1-DliAryrR7_kQMe2tKSZ29i_0TEwdNL7A/s1600/age.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" height="400" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgKdiOpX1eV-hd0mJwlnVQU0jcvtE8_0WaA35qw2JNFRpZ_H0lDVxB4jDXQqVup85YBdb7Yzb62kOa-jH4aU_8OgzOQfcGl_ZoGv1ygl1w1-1-DliAryrR7_kQMe2tKSZ29i_0TEwdNL7A/s400/age.png" width="400" /></a></div>
<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjBkwy7B-FVmQWeSJhngu0IjcSft3w9twVJA2DqNomKtH63nUzLKuolpfvZJ7wvH54UnA-mcBOZcx38UBKy_5L7ZnBuTyKxeMCbAqZohhkWMyi4Mjh9gIVGIL4QVIvphzshorhoHHMnUNQ/s1600/race.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" height="400" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjBkwy7B-FVmQWeSJhngu0IjcSft3w9twVJA2DqNomKtH63nUzLKuolpfvZJ7wvH54UnA-mcBOZcx38UBKy_5L7ZnBuTyKxeMCbAqZohhkWMyi4Mjh9gIVGIL4QVIvphzshorhoHHMnUNQ/s400/race.png" width="400" /></a><a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjBiQG66WIjQ4WA1hVS0em1H08DB6NbNAEXAwQtXF2iA9LqvVXI8BoqPgwSqK3qlH0LxVbakBXL8tNwx2bedCO5D0xlBr40wBGzauKWVTDIokFEpuWl8EKx30rNa_ewuBWs8W7RO0C-b60/s1600/state.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" height="400" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjBiQG66WIjQ4WA1hVS0em1H08DB6NbNAEXAwQtXF2iA9LqvVXI8BoqPgwSqK3qlH0LxVbakBXL8tNwx2bedCO5D0xlBr40wBGzauKWVTDIokFEpuWl8EKx30rNa_ewuBWs8W7RO0C-b60/s400/state.png" width="400" /> </a></div>
<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhXcUarg15V0E_lY0YMaUWS2fNlCsdv0TXpadKxfEHRJSGk5DuA2BzPdhp0gS7SDrymd4gJ9CwZfUSaHe1KbO_stIOI-S-5BKg16G9n4s6FbHJUHg6v1TXlP4iVDMWTZeKN4bVOFx1xzro/s1600/yearsconf.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" height="400" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhXcUarg15V0E_lY0YMaUWS2fNlCsdv0TXpadKxfEHRJSGk5DuA2BzPdhp0gS7SDrymd4gJ9CwZfUSaHe1KbO_stIOI-S-5BKg16G9n4s6FbHJUHg6v1TXlP4iVDMWTZeKN4bVOFx1xzro/s400/yearsconf.png" width="400" /></a></div>
<div class="separator" style="clear: both; text-align: center;">
<br /></div>
<div class="separator" style="clear: both; text-align: center;">
<br /></div>
Here is the R code used to create the above pie charts.<br />
<br />
<br />
<span style="font-family: "Courier New",Courier,monospace;"># National Registry of Exonerations <br /># pie charts<br /><br />library(XML)<br /><br />u <- "http://www.law.umich.edu/special/exoneration/Pages/detaillist.aspx"<br /><br />listu <- readHTMLTable(u)<br /><br />exondf <- listu[[7]]<br />data <- exondf[24:nrow(exondf),]<br />names(data) <- as.character(unlist(exondf[4,]))<br /><br /># transform data<br />data$Age <- droplevels(data$Age)<br />data$Race <- droplevels(data$Race)<br />data$State <- droplevels(data$State)<br />data$Crime <- droplevels(data$Crime)<br />data$Sentence <- droplevels(data$Sentence)<br />data$Convicted <- droplevels(data$Convicted)<br />data$Exonerated <- droplevels(data$Exonerated)<br /><br />data$AgeCNV <- as.numeric(as.character(data$Age))<br />data$ConvictedCNV <- as.numeric(as.character(data$Convicted))<br />data$ExoneratedCNV <- as.numeric(as.character(data$Exonerated))<br /><br />data$AgeCNV_floor <- floor(data$AgeCNV/10)*10<br />data$ConfinedYrs <- data$ExoneratedCNV - data$ConvictedCNV<br />data$ConfinedYrs_floor <- floor(data$ConfinedYrs/5)*5<br /><br /># plot pie charts<br /><br />LABELS <- c("10-19","20-29","30-39","40-49","50-59","60-69","")<br />pie(table(data$AgeCNV_floor), labels=LABELS, main="Age Exonerated")<br /><br />pie(table(data$Race), main="Race")<br /><br />pie(tail(sort(table(data$State)),10), main="Top 10 States")<br /><br />LABELS <- c("0-4","5-9","10-14","15-19","20-24","25-29","30-34","35+")<br />pie(table(data$ConfinedYrs_floor), labels=LABELS, main="Years Confined")</span>larrydaghttp://www.blogger.com/profile/06592231655463233562noreply@blogger.com4tag:blogger.com,1999:blog-4196841648700777718.post-9701615489113409752012-04-04T06:06:00.003-07:002012-04-04T06:06:23.087-07:00Google Scholar MetricsGoogle Scholar, the Google produced search method for scholarly journals and publications, has a new way of tracking publication metrics. <a href="http://googlescholar.blogspot.com/2012/04/google-scholar-metrics-for-publications.html">Google Scholar Metrics for publications</a> gives an indexed look at the publishers and figure out which publishers are cited the most. Google Scholar Metrics will allow the searcher to find publishers that are well respected in any given field of research. As an example here are the <a href="http://scholar.google.com/citations?view_op=top_venues">Top 100 Google Scholar publishers</a>.<br />
<br />
So I naturally want to see what Google thinks of some of the fields that I find interesting and useful.<br />
<br /><a href="http://scholar.google.com/citations?hl=en&view_op=search_venues&vq=[operations+research]">Operations Research</a><br />
<ol>
<li> European Journal of Operational Research</li>
<li> Computers & Operations Research</li>
<li> Operations Research</li>
<li> Journal of the Operational Research Society</li>
<li> Annals of Operations Research</li>
</ol>
<a href="http://scholar.google.com/citations?hl=en&view_op=search_venues&vq=[machine+learning]">Machine Learning </a><br />
<ol>
<li>The Journal of Machine Learning Research</li>
<li>Annual International Conference on Machine Learning</li>
<li>Machine Learning</li>
<li>European Conference on Machine learning and knowledge discovery in databases</li>
<li>International Conference on Machine Learning and Cybernetics</li>
</ol>
<a href="http://scholar.google.com/citations?hl=en&view_op=search_venues&vq=[applied+statistics]">Applied Statistics </a><br />
<ol>
<li>The Annals of Applied Statistics</li>
<li> Journal of the Royal Statistical Society: Series C (Applied Statistics)</li>
<li> Journal of Applied Statistics</li>
<li> QUALITY CONTROL AND APPLIED STATISTICS</li>
<li> International Journal of Applied Mathematics and Statistics</li>
</ol>
<a href="http://scholar.google.com/citations?hl=en&view_op=search_venues&vq=[management+science]">Management Science</a> <br />
<ol>
<li> Management Science</li>
<li> Pest Management Science (????)</li>
<li> Health Care Management Science</li>
<li> Conflict Management and Peace Science (????)</li>
<li> Computational Management Science</li>
</ol>
<table class="cit-table"><tbody>
<tr class="cit-table item"><td id="col-pos"><br /></td><td id="col-vtitle"><br /></td><td id="col-hindex"><br /></td><td id="col-hmedian">The Management Science category looks like it needs a lot of work. I didn't know there were so many other forms of Management Science. I guess that it is too generic of a term and the age old debate continues.</td></tr>
<tr class="cit-table item"><td id="col-pos"><br /></td><td id="col-vtitle"><br /></td><td id="col-hindex"><br /></td><td id="col-hmedian"><br /></td></tr>
<tr class="cit-table item"><td id="col-pos"><br /></td><td id="col-vtitle"><br /></td><td id="col-hindex"><br /></td><td id="col-hmedian"><br /></td></tr>
<tr class="cit-table item"><td id="col-pos"><br /></td><td id="col-vtitle"><br /></td><td id="col-hindex"><br /></td><td id="col-hmedian"><br /></td></tr>
<tr class="cit-table item"><td id="col-pos"><br /></td><td id="col-vtitle"><br /></td><td id="col-vtitle"><br /></td><td id="col-vtitle"><br /></td><td id="col-vtitle"><br /></td><td id="col-vtitle"><br /></td><td id="col-vtitle"><br /></td><td id="col-vtitle"><br /></td><td id="col-vtitle"><br /></td><td id="col-vtitle"><br /></td><td id="col-vtitle"><br /></td><td id="col-vtitle"><br /></td></tr>
</tbody></table>larrydaghttp://www.blogger.com/profile/06592231655463233562noreply@blogger.com0tag:blogger.com,1999:blog-4196841648700777718.post-14272798715211557182012-03-15T03:21:00.002-07:002012-03-15T03:24:14.941-07:00OpenOpt Suite 0.38<div class="entry-content"> <p>I'm glad to inform you about new OpenOpt Suite quarter release 0.38, free (BSD license) Python-written software:</p><p><a href="http://openopt.org/OOFramework">OpenOpt</a>:</p><p> <a href="http://openopt.org/interalg">interalg</a> can handle discrete variables (see <a href="http://openopt.org/MINLP">MINLP</a> for examples)<br /> interalg can handle multiobjective problems (<a href="http://openopt.org/MOP">MOP</a>)<br /> interalg can handle problems with parameters fixedVars/freeVars<br /> Many interalg improvements and some bugfixes<br /> Add another <a href="http://openopt.org/EIG">EIG</a> solver: numpy.linalg.eig<br /> New <a href="http://openopt.org/LLSP">LLSP</a> solver pymls with box bounds handling </p><p><a href="http://openopt.org/FuncDesigner">FuncDesigner</a>:</p><p> Some improvements for sum()<br /> Add funcs tanh, arctanh, arcsinh, arccosh<br /> Can solve <a href="http://openopt.org/EIG">EIG</a> built from derivatives of several functions, obtained by automatic differentiation by FuncDesigner</p><p><a href="http://openopt.org/SpaceFuncs">SpaceFuncs</a>:</p><p> Add method point.symmetry(Point|Line|Plane)<br /> Add method LineSegment.middle<br /> Add method Point.rotate(Center, angle) </p><p><a href="http://openopt.org/DerApproximator">DerApproximator</a>:</p><p> Minor changes</p><p>See also: <a href="http://openopt.org/FuturePlans">FuturePlans</a>.</p><p>Regards, D.</p> </div>Dmitreyhttp://www.blogger.com/profile/18259817544023299492noreply@blogger.com1tag:blogger.com,1999:blog-4196841648700777718.post-8203775689903199222012-02-02T07:48:00.000-08:002012-02-02T07:48:36.693-08:00R graphic used for Facebook IPO<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhY13Zv7cwq6ChoxKZ-OM5rDr9XPlahUw6Is2K15sEZlhWorWk65QgbPJ6N5QVCKoJfVmNLQsJn7qeuEKff_uqpR6vXUF7Ak1w2o2YsB2haeOdFqS_fPaf0awZo_w4l5BUnHBcm7t-Y4QE/s1600/facebook_graph.png" imageanchor="1" style="clear: left; float: left; margin-bottom: 1em; margin-right: 1em;"><img border="0" height="99" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhY13Zv7cwq6ChoxKZ-OM5rDr9XPlahUw6Is2K15sEZlhWorWk65QgbPJ6N5QVCKoJfVmNLQsJn7qeuEKff_uqpR6vXUF7Ak1w2o2YsB2haeOdFqS_fPaf0awZo_w4l5BUnHBcm7t-Y4QE/s200/facebook_graph.png" width="200" /></a></div>
Apparently former Facebook intern, Paul Butler, graphic of the Facebook social network graph is being used for <a href="http://mashable.com/2012/02/01/facebook-files-ipo/">Facebook's IPO</a>. The social network graphic is featured on Page 7 of the IPO filing. His graphic was featured on <a href="http://mashable.com/2010/12/13/facebook-members-visualization/">mashable</a> and <a href="http://www.r-bloggers.com/facebooks-social-network-graph/">R-bloggers</a> not too long ago. The graphic is of Facebook connections between city centers around the world. Paul used an ingenious method of color transparency and great circle arcs to display the social network graph.<br />
<br />
This is just one of the really cool things you can do with R. Not only is R used as a visual medium but also to calculate the great circle paths. This is really neat to see R in such a high profile setting. If you want to learn more about R you can read an IEORTools post about <a href="http://industrialengineertools.blogspot.com/2010/10/r-links-for-beginner-on-world.html">R links for beginners on World Statistics Day</a>. Also there are many books that you can buy on <a href="http://www.ieortools.com/r-project/">R programming</a> at the <a href="http://www.ieortools.com/">IEORTools</a> Online Store.larrydaghttp://www.blogger.com/profile/06592231655463233562noreply@blogger.com7tag:blogger.com,1999:blog-4196841648700777718.post-66272445733336387402012-01-12T06:13:00.000-08:002012-01-12T06:13:16.476-08:00Should science be open<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiFlUAatIvgrCLPxnOmc8paYZ33-hn9I5e6BvIneOGjrqICwHEGzjTULLuJhij6cdG6h_ikShKn4X4SY5qUIikIgf-AZP0iibXYa0s25uE9G5vxjpOwTG0Sr6wVbEFNkjOpR_wI8QxIqdg/s1600/science_viewCN_0156.jpg" imageanchor="1" style="clear: left; float: left; margin-bottom: 1em; margin-right: 1em;"><img border="0" height="134" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiFlUAatIvgrCLPxnOmc8paYZ33-hn9I5e6BvIneOGjrqICwHEGzjTULLuJhij6cdG6h_ikShKn4X4SY5qUIikIgf-AZP0iibXYa0s25uE9G5vxjpOwTG0Sr6wVbEFNkjOpR_wI8QxIqdg/s200/science_viewCN_0156.jpg" width="200" /></a></div>
Two interesting articles appeared this week in some blogs I frequent about technology and science. The first is an Op-ed in the New York Times titled <a href="http://www.nytimes.com/2012/01/11/opinion/research-bought-then-paid-for.html?_r=1">Research Bought, Then Paid For</a> and the next is <a href="http://lemire.me/blog/archives/2012/01/10/open-science-is-hard/">Open Science: why is it so hard?</a> The two articles are a different take on the idea that scientific findings should be open for everyone. Someone who is outside the scientific community might think that statement is silly. Of course science is open. No one has a copyright or a monopoly on scientific or mathematical discoveries. Yet that is not the real issue. The real issue is the access to those scientific discoveries. In some cases the scientific discoveries are paid for by public subsidies.<br />
<br />
The main focus of those two articles is that science has been hijacked by the publishers. The articles even go so far as saying the hijacking is a monopoly of sorts. I think monopoly is too strong of an analogy but the publishers do have a lot of control. The control is mostly about access to the science. The publishers own the copyright and can limit access to anyone unless a fee is paid. A lot of the times these fees are rather high. Now it looks like with the <a href="http://en.wikipedia.org/wiki/Research_Works_Act">Research Works Act</a> the access to publicly funded scientific research will be limited as well. Access to the science is the crux of the debate.<br />
<br />
Academics rely on publishing of their scientific findings for further funding of their research. It is part of the academic circle of life. Publishing begets more funding which begets more publishing and the cycle continues. I do believe academic community deserves to get compensated for their research. I'm not sure how much residual income they get other than peer review notoriety from their published content. Publishers seem, again, to have a lot of the control. <br />
<br />
I am not an academic researcher. My work is trying to help organizations better themselves by using the learning, skills, and knowledge I have acquired through the years as an Operations Research professional. I try to keep up to date on the latest research and methods by studying journals, networking with colleagues, and reading articles. I rely on scientific access quite a bit in staying up to date with the latest findings. I rely on the academic community so I can improve my knowledge and skills. Yet it seems very difficult for my to gain access to a lot of good research. There has to be a common ground for access to the science. I wish I had a simple solution to this issue but it seems very large and very complicated. There are a lot of interactions that I am sure I am glossing over. Yet I am a big fan of the idea of <a href="http://en.wikipedia.org/wiki/Open_science">Open Science</a>.<br />
<br />
There are some publishers that do understand this problem. <a href="http://www.informs.org/">INFORMS</a> seems to get this issue rather well. They do not charge a lot for their journals. In fact as part of membership INFORMS allows two free subscriptions to any journal of your choosing. In addition to that the <a href="http://www.informs.org/Find-Research-Publications/Journals">PubsOnLine Suite</a> is available for $99 which is 12 journals for a whole year. That is a bargain compared to some other publishers. So not all publishers are pure evil. There are some good ones.larrydaghttp://www.blogger.com/profile/06592231655463233562noreply@blogger.com11tag:blogger.com,1999:blog-4196841648700777718.post-22877730835648112582012-01-02T14:19:00.000-08:002012-01-02T14:21:29.049-08:00IEORTools.com Resources addedI've decided to spruce up my personal website <a href="http://www.ieortools.com/">IEORTools.com</a>. I want to add some additional resources to it along with the book store. Most of the content will be relevant reference links to Industrial Engineering and Operations Research professionals.<br />
<br />
The first thing I did was added a Resources side menu. The Resources side menu will link to relevant resource sections. So far I have created the following resources<br />
<ul>
<li><a href="http://www.ieortools.com/optimization/">Optimization</a></li>
<li><a href="http://www.ieortools.com/statistics-1037466943/">Statistics</a></li>
<li><a href="http://www.ieortools.com/data-mining-841068338/">Data Mining</a></li>
</ul>
These links are a collection of resources that I have accumulated over the years. The links are a great reference and hopefully I can build them up more. I'm going to be creating more content on ieortools.com site as opposed to the blog because I'm just running out of room.larrydaghttp://www.blogger.com/profile/06592231655463233562noreply@blogger.com3