https://www.datacamp.com/community/tutorials/preprocessing-in-data-science-part-1-centering-scaling-and-knn

This article will explain the importance of preprocessing in the machine learning pipeline by examining how centering and scaling can improve model performance.

Preprocessing in Data Science (Part 1)

Preprocessing in Data Science (Part 1): Centering, Scaling, and KNN

Nowadays, the data science field is hot, and it is unlikely that this will change in the near future. While a data driven approach is finding its way into all facets of business, companies are fiercely fighting for the best data analytic skills that are available in the market, and <a href="http://www.kdnuggets.com/2015/01/predictions-2015-analytics-data-science-hiring-market.html" target="_blank">salaries for data science roles are going in overdrive</a>.Companies&rsquo; increased focus on acquiring data science talent goes hand in hand with the creation of a whole new set of <a href="http://www.forbes.com/sites/rawnshah/2015/10/06/data-science-falls-into-many-roles/" target="_blank">data science roles and titles</a>. Sometimes new roles and titles are added to reflect changing needs; other times they are probably created as a creative way to differentiate from fellow recruiters. Either way, it&rsquo;s hard to get a general understanding of the different job roles, and it gets even harder when you&rsquo;re looking to figure out what job role could best fit your personal and professional ambitions.

<h2>The Data Science Industry: Who Does What</h2>

<a href="https://www.datacamp.com/?utm_source=blog&amp;utm_medium=blog&amp;utm_campaign=infographic_data_science_roles" target="_blank">DataCamp</a> took a look at this avalanche of data science job postings in an attempt to unravel these cool-sounding and playful job titles into a comparison of different data science related careers. We summarized the results in our latest infographic &ldquo;The Data Science Industry: Who Does What&rdquo;:<a href="http://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1523010085/main-qimg-727dcab957de412c83fcc37f382233ee_z8rmxy.png"><img alt="Data Science Jobs Infographic" class="aligncenter size-full wp-image-698" height="9921" src="http://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1523010085/main-qimg-727dcab957de412c83fcc37f382233ee_z8rmxy.png" width="1701" /></a>In this infographic we compare the roles of data scientists, data analysts, data architects, data engineers, statisticians and many more. We have a look at their roles within companies and the data science process, what technologies they have mastered, and what the typical skillset and mindset is for each role. Furthermore, we look at the top employers that are currently hiring these different data science roles and how the average national salaries of these roles map out against each other.Hopefully this infographic will help you to better understand the different job roles that are available to data passionate professionals. If you want to learn more on data science, make sure to check out <a href="https://www.datacamp.com/?utm_source=blog&amp;utm_medium=blog&amp;utm_campaign=infographic_data_science_roles">DataCamp&rsquo;s interactive R and data science tutorials</a> and join 220,000 data enthusiasts!Interested in some of our other infographics? Check out <a href="https://www.datacamp.com/community/tutorials/statistical-language-wars-the-infograph/" target="_blank" title="R vs Sas vs Spss for data analysis">Statistical Language Wars</a>, <a href="https://www.datacamp.com/community/tutorials/how-to-become-a-data-scientist/" target="_blank" title="How to become a data scientist">How to Become a Data Scientist in 8 Easy Steps</a> and <a href="https://www.datacamp.com/community/tutorials/r-or-python-for-data-analysis/" target="_blank">R vs Python.</a>&nbsp;&nbsp;

https://www.datacamp.com/community/tutorials/data-science-industry-infographic

This infograph compares the roles of data scientists, data analysts, data architects, data engineers and more in the data science industry.

The Data Science Industry: Who Does What (Infographic)

When you need to do web scraping, you would normally make use of Hadley Wickham&rsquo;s <code class="lang-r">rvest</code> package. This package provides an easy to use, out of the box solution to fetch the html code that generates a webpage. However, when the website or webpage makes use of JavaScript to display the data you&#39;re interested in, the <code class="lang-r">rvest</code> package misses the required functionality. One solution is to make use of PhantomJS.

(Want to practice importing more data into R?&nbsp;Try <a href="https://www.datacamp.com/courses/importing-data-in-r-part-1/"> this tutorial</a>.)

<h2>Content</h2>

<nav>
<ul>
	<li><a href="#load">Load the necessary packages</a></li>
	<li><a href="#scraping">Scraping Javascript Generated Data with R</a></li>
	<li><a href="#conclusion">Conclusion</a></li>
</ul>
</nav>
<div id="scoped-content">
<style>
:target:before {
content:"";
display:block;
height:150px; 
margin:-150px 0 0; 
}
</style>
 

<h2 id="load">Load the necessary packages</h2>

As a first step, you&#39;ll have to load all packages that we need for this analysis into the workspace (if you haven&#39;t installed these packages on your local system yet, use <code class="lang-r">install.packages()</code> to make them available):

<pre class="pre-no-padding">
<code class="lang-r" class="r">library(rvest)
library(stringr)
library(plyr)
library(dplyr)
library(ggvis)
library(knitr)
options(digits = 4)
</code></pre>

<h2 id="scraping">Scraping Javascript Generated Data with R</h2>

The next step is the collection of the TechStars data using PhantomJS. Check out the following basic .js file:

<pre class="pre-no-padding">
<code class="lang-r" class="js">// scrape_techstars.js

var webPage = require('webpage');
var page = webPage.create();

var fs = require('fs');
var path = 'techstars.html'

page.open('http://www.techstars.com/companies/stats/', function (status) {
 var content = page.content;
 fs.write(path,content,'w')
 phantom.exit();
});
</code></pre>

The script basically renders the HTML page after the underlying javascript code has done its work, allowing you to fetch the HTML page, with all the tables in there. To stay in R for the rest of this analysis, we suggest you use the <code class="lang-r">system()</code> function to invoke PhantomJS (you&#39;ll have to <a href="http://phantomjs.org/download.html">download and install PhantomJS</a> and put it in your working directory):

<pre class="pre-no-padding">
<code class="lang-r" class="r"># Let phantomJS scrape techstars, output is written to techstars.html
system("./phantomjs scrape_techstars.js")
</code></pre>

After this small detour, you finally have an HTML file, <code class="lang-r">techstars.html</code>, on our local system, that can be scrape with <code class="lang-r">rvest</code>. An inspection of the Techstars webpage reveals that the tables we&#39;re interested in are located in divs with the css class <code class="lang-r">batch</code>:

<pre class="pre-no-padding">
<code class="lang-r" class="r">batches &lt;- html("techstars.html") %&gt;%
 html_nodes(".batch")

class(batches)
</code></pre>

<pre class="pre-no-padding">
<code class="lang-r" class="r">[1] "XMLNodeSet"
</code></pre>

You now have a list of <code class="lang-r">XMLNodeSet</code> objects: each object contains the data for a single TechStars batch. In there, we can find information concerning the batch location, the year, the season, but also about the companies, their current headquarters, their current status and the amount of funding they raised in total. We will not go into detail on the data collection and cleaning steps below; you can execute the code yourself and inspect what they accomplish. You&#39;ll see that some custom cleaning is going on to make sure that each bit of information is nicely formatted:

<pre class="pre-no-padding">
<code class="lang-r" class="r">batch_titles &lt;- batches %&gt;%
 html_nodes(".batch_class") %&gt;%
 html_text()

batch_season &lt;- str_extract(batch_titles, "(Fall|Spring|Winter|Summer)")
batch_year &lt;- str_extract(batch_titles, "([[:digit:]]{4})")
# location info is everything in the batch title that is not year info or season info
batch_location &lt;- sub("\\s+$", "",
 sub("([[:digit:]]{4})", "",
 sub("(Fall|Spring|Winter|Summer)","",batch_titles)))

# create data frame with batch info.
batch_info &lt;- data.frame(location = batch_location,
 year = batch_year,
 season = batch_season)

breakdown &lt;- lapply(batches, function(x) {
 company_info &lt;- x %&gt;% html_nodes(".parent")
 companies_single_batch &lt;- lapply(company_info, function(y){
 as.list(gsub("\\[\\+\\]\\[\\-\\]\\s", "", y %&gt;%
 html_nodes("td") %&gt;%
 html_text()))
 })
 df &lt;- data.frame(matrix(unlist(companies_single_batch),
 nrow=length(companies_single_batch),
 byrow=T,
 dimnames = list(NULL, c("company","funding","status","hq"))))
 return(df)
})

# Add batch info to breakdown
batch_info_extended &lt;- batch_info[rep(seq_len(nrow(batch_info)),
 sapply(breakdown, nrow)),]
breakdown_merged &lt;- rbind.fill(breakdown)

# Merge all information
techstars &lt;- tbl_df(cbind(breakdown_merged, batch_info_extended)) %&gt;%
 mutate(funding = as.numeric(gsub(",","",gsub("\\$","",funding))))
</code></pre>

With a combination of core R, <code class="lang-r">rvest</code>, <code class="lang-r">plyr</code> and <code class="lang-r">dplyr</code> functions, we now we have the <code class="lang-r">techstars</code> data frame; a data set of all TechStars company, with all publicly available information that is nicely formatted:

<pre class="pre-no-padding">
<code class="lang-r" class="r">techstars
</code></pre>

<pre class="pre-no-padding">
<code class="lang-r">## Source: local data frame [535 x 7]
##
## company funding status hq location year season
## 1 Accountable 110000 Active Fort Worth, TX Austin 2013 Fall
## 2 Atlas 1180000 Active Austin, TX Austin 2013 Fall
## 3 Embrace 110000 Failed Austin, TX Austin 2013 Fall
## 4 Filament Labs 1490000 Active Austin, TX Austin 2013 Fall
## 5 Fosbury 300000 Active Austin, TX Austin 2013 Fall
## 6 Gone! 840000 Active San Francisco, CA Austin 2013 Fall
## 7 MarketVibe 110000 Acquired Austin, TX Austin 2013 Fall
## 8 Plum 1630000 Active Austin, TX Austin 2013 Fall
## 9 ProtoExchange 110000 Active Austin, TX Austin 2013 Fall
## 10 Testlio 1020000 Active Austin, TX Austin 2013 Fall
## .. ... ... ... ... ... ... ...
</code></pre>

<pre class="pre-no-padding">
<code class="lang-r" class="r">names(techstars)
</code></pre>

<pre class="pre-no-padding">
<code class="lang-r" class="r">## [1] "company" "funding" "status" "hq" "location" "year"
## [7] "season"
</code></pre>

<h2 id="conclusion">Conclusion</h2>

Using the above combination of tools and code, we managed to scrape data from a website that uses a JavaScript script to generate its data. As one can see, this is a very structured process, that can be easily done once the initial code is available. Want to learn R, and looking for data science and R tutorials and courses? Check out <a href="https://www.datacamp.com/utm_source=web_scraping_blog&amp;utm_medium=blog&amp;utm_campaign=web_scraping_blog/">DataCamp</a>
</div>

https://www.datacamp.com/community/tutorials/scraping-javascript-generated-data-with-r

Short tutorial on scraping Javascript generated data with R using PhantomJS.

Short R tutorial: Scraping Javascript Generated Data with R

Web Scraping with R and PhantomJS

Data Science has its own language. So, if you want to have at least a slight chance of surviving in the enterprise world of tomorrow -with its obsessive focus on collecting and analyzing data- you better have started yesterday with learning this terminology.


Luckily, and inspired by the <a href="http://techcrunch.com/2014/12/15/how-to-speak-startup/" target="_blank">&ldquo;How To Speak Startup&rdquo;</a> article on TechCrunch, the online data science school&nbsp;<a href="https://www.datacamp.com/?utm_source=How%20to%20speak%20data%20science&amp;utm_medium=blog&amp;utm_campaign=How%20to%20speak%20data%20science" target="_blank" title="Online R and Data Science Training">DataCamp</a> is here to help you with a great list of often-heard&nbsp;data science terminology and quotes:

(Note: This is an attempt to be funny. Whether this is a successful attempt is an opinion and open for debate. That it shouldn&rsquo;t be taken serious nor offensive is a fact.)
 
<ul>
<li>Data scientist - Average programmer looking for a job that pays as much as what a top programmer would get. Sometimes also goes by the name &quot;data analyst&quot;. <a href="http://ctt.ec/fd39T">Click To Tweet.</a></li>

<li>Statistician - Mathematician who can&rsquo;t program. <a href="http://ctt.ec/5GeK0">Click To Tweet</a></li>

<li>&quot;Our company is big data ready&quot; - My software vendor has done a great up-sell. <a href="http://ctt.ec/e60r2">Click To Tweet</a></li>

<li>&quot;We measure everything&quot; - We have absolutely no clue what to measure. <a href="http://ctt.ec/agQBd">Click To Tweet</a></li>

<li>Data savvy manager - Title used by managers active in marketing, sales or HR whom put pie charts in their powerpoint presentation.<a href="http://ctt.ec/yb5CO"> Click To Tweet</a></li>

<li>&quot;Correlation does not imply causation&quot; - We looked at the wrong data set and can&rsquo;t draw any conclusions from it. Often represented in a graph to create the illusion of adding value. <a href="http://ctt.ec/bYG2e">Click To Tweet</a></li>

<li>Machine Learning - Statistical technique used by the sales and marketing department of big data vendors to secure their yearly bonus. (also see &quot;Our company is big data ready&quot;) <a href="http://ctt.ec/V03xb">Click To Tweet</a></li>

<li>Chief Data Scientist (CDS) - Former CTO (also see data scientist) <a href="http://ctt.ec/aIF5E">Click To Tweet</a></li>

<li>Hadoop - Open-source software used for distributed computing. Data Scientists seem to have a quota to drop the name every two sentences when talking big data, but most only know the logo is a yellow elephant.&nbsp;<a href="http://ctt.ec/SK6Z6">Click To Tweet</a></li>

<li>&ldquo;Being a data scientist is the sexiest job in the 21st century&quot; - Although a quote often used in the data science industry, statistical underpinned proof remains missing. <a href="http://ctt.ec/R5tvh">Click To Tweet</a></li>

<li>Data science bootcamp - Headhunting firm marketing itself as a school. <a href="http://ctt.ec/2p735">Click To Tweet</a></li>

<li>&ldquo;We booked these results with a small sample&quot; - Our financial budget wasn&rsquo;t large enough to perform a statistical significant data analysis.
<a href="http://ctt.ec/8Ura6">Click To Tweet</a></li>

<li>&ldquo;We implemented a data-driven decision making process&rdquo; - In the past we didn&rsquo;t have a clue what we were doing. <a href="http://ctt.ec/1cPGk">Click To Tweet</a></li>

<li>&quot;We use cutting edge predictive modeling techniques to forecast our results&rdquo; - We run a linear regression model and then ignore the result. <a href="http://ctt.ec/BacF3">Click To Tweet</a></li>

<li>&quot;There is a significant effect but ...&quot; - Sentence-start used by data scientists or statisticians when they&rsquo;ve put weeks of work into their analysis, the results look fishy and not as expected, and there is no time to redo the analysis. <a href="http://ctt.ec/UXfJd">Click To Tweet</a></li>
</ul>
 
Hope this was helpful! Good Luck.

https://www.datacamp.com/community/tutorials/how-to-speak-data-science

Data Science has its own language. This is a great list of often-heard data science terminology and quotes by Datacamp.

How to Speak Data Science

This data.table R tutorial&nbsp;explains the basics of the <code>DT[i, j, by]</code> command which is core to the data.table package. If you want to learn more on the data.table package, <a href="https://www.datacamp.com/courses/data-analysis-the-data-table-way/?utm_source=blog&amp;utm_medium=blog&amp;utm_content=data.table%20i-j-by%20blog&amp;utm_campaign=data.table%20i-j-by%20blog" target="_blank" title="data.table R tutorial">DataCamp </a>provides an <a href="https://www.datacamp.com/courses/data-analysis-the-data-table-way/?utm_source=blog&amp;utm_medium=blog&amp;utm_content=data.table%20i-j-by%20blog&amp;utm_campaign=data.table%20i-j-by%20blog" target="_blank" title="data.table r tutorial">interactive R course on the data.table package</a>. The course has more than 35&nbsp;interactive R exercises - all taking place in the comfort of your own browser - and several videos with&nbsp;<a href="https://twitter.com/MattDowle" target="_blank">Matt </a><a href="https://www.youtube.com/watch?v=qLrdYhizEMg" target="_blank">Dowle</a>, main author of the data.table package, and <a href="https://twitter.com/arun_sriniv" target="_blank">Arun Srinivasan</a>, major contributor. <a href="https://www.datacamp.com/courses/data-analysis-the-data-table-way/?utm_source=blog&amp;utm_medium=blog&amp;utm_content=data.table%20i-j-by%20blog&amp;utm_campaign=data.table%20i-j-by%20blog" target="_blank">Try if for free</a>.
<a href="http://blog.datacamp.com/wp-content/uploads/2014/10/learning-exp-matt-video.png"><img alt="A data.table R tutorial by DataCamp" class="alignnone wp-image-405 size-large" height="154" src="https://s3.amazonaws.com/assets.datacamp.com/blog_assets/data.table+Tutorial/learning-exp-matt-video-1024x334.png" width="474" /></a> 
If you have already worked with large datasets in RAM (1 to more than 100GB), you know that a <code>data.frame</code> can be limiting: the time it takes to do certain things is just too long. Data.table solves this for you by reducing computing time. Evenmore, it also makes it easier to do more with less typing. Once you master the data.table&nbsp;syntax from this data.table R tutorial, the simplicity of doing complicated operations will astonish you. So you will not only be reducing computing time, but programming time as well.
The <code>DT[i,j,by]</code> command has three parts: <code>i</code>, <code>j</code> and <code>by</code>. If you think in SQL terminology, the <code>i</code> corresponds to WHERE, <code>j</code> to SELECT and <code>by</code> to GROUP BY. We talk about the command by saying &ldquo;Take DT, subset the rows using &lsquo;<code>i</code>&rsquo;, then calculate &lsquo;<code>j</code>&rsquo; grouped by &lsquo;<code>by</code>&rsquo;&rdquo;. So in a simple example and using the <code>hflights</code> dataset (so you can reproduce all the examples) this gives:
<pre><code>
library(hflights)
library(data.table)
DT &amp;lt;- as.data.table(hflights)
DT[Month==10,mean(na.omit(AirTime)), by=UniqueCarrier]
</code></pre>
<pre><code>
UniqueCarrier V1
AA 68.76471
AS 255.29032
B6 176.93548
CO 141.52861
...
</code></pre>
Where we subsetted the data table to keep only the rows of the 10th Month of the year, calculated the average AirTime of the planes that actually flew (that&rsquo;s why <code>na.omit()</code> is used, cancelled flights don&#39;t have a value for their AirTime) and then grouped the results by their Carrier. We can see for example that AA (American Airlines) has a very short average AirTime compared to AS (Alaska Airlines). &nbsp;Did you also notice that R base functions can be used in the <code>j</code> part? We will get to that later.
<h2>The i part</h2>
The &lsquo;i&rsquo; part is used for subsetting on rows, just like in a data frame.
<pre><code>
DT[2:5]
</code></pre>
<pre><code>
#selects the second to the fifth row of DT
Year Month DayofMonth DayOfWeek DepTime ArrTime UniqueCarrier FlightNum TailNum ActualElapsedTime AirTime
2011 1 2 7 1401 1501 AA 428 N557AA 60 45
2011 1 3 1 1352 1502 AA 428 N541AA 70 48
2011 1 4 2 1403 1513 AA 428 N403AA 70 39
2011 1 5 3 1405 1507 AA 428 N492AA 62 44

ArrDelay DepDelay Origin Dest Distance TaxiIn TaxiOut Cancelled CancellationCode Diverted
-9 1 IAH DFW 224 6 9 0 0
-8 -8 IAH DFW 224 5 17 0 0
 3 3 IAH DFW 224 9 22 0 0
-3 5 IAH DFW 224 9 9 0 0
</code></pre>

But you can also use column names, as they are evaluated in the scope of DT.
<pre><code>
DT[UniqueCarrier==&amp;quot;AA&amp;quot;]
</code></pre>
<pre><code>
#Returns all those rows where the Carrier is American Airlines
Year Month DayofMonth DayOfWeek DepTime ArrTime UniqueCarrier FlightNum TailNum ActualElapsedTime
2011 1 1 6 1400 1500 AA 428 N576AA 60
2011 1 2 7 1401 1501 AA 428 N557AA 60
2011 1 3 1 1352 1502 AA 428 N541AA 70
2011 1 4 2 1403 1513 AA 428 N403AA 70
2011 1 5 3 1405 1507 AA 428 N492AA 62
---
2011 12 27 2 1021 1333 AA 2234 N3ETAA 132
2011 12 28 3 1015 1329 AA 2234 N3FJAA 134
2011 12 29 4 1023 1335 AA 2234 N3GSAA 132
2011 12 30 5 1024 1334 AA 2234 N3BAAA 130
2011 12 31 6 1024 1343 AA 2234 N3HNAA 139
AirTime ArrDelay DepDelay Origin Dest Distance TaxiIn TaxiOut Cancelled CancellationCode Diverted
40 -10 0 IAH DFW 224 7 13 0 0
45 -9 1 IAH DFW 224 6 9 0 0
48 -8 -8 IAH DFW 224 5 17 0 0
39 3 3 IAH DFW 224 9 22 0 0
44 -3 5 IAH DFW 224 9 9 0 0
---
112 -12 1 IAH MIA 964 8 12 0 0
112 -16 -5 IAH MIA 964 9 13 0 0
110 -10 3 IAH MIA 964 12 10 0 0
110 -11 4 IAH MIA 964 9 11 0 0
119 -2 4 IAH MIA 964 8 12 0 0
</code></pre>

Notice that you don&rsquo;t have to use a comma for subsetting rows in a data table. In a data.frame doing this <code>DF[2:5]</code> would give all the rows of the 2nd to 5th column. Instead (as everyone reading this obviously knows), we have to specify <code>DF[2:5,]</code>. Also notice that <code>DT[,2:5]</code> does not mean anything for data tables, as is explained in the first question of the <a href="http://cran.r-project.org/web/packages/data.table/vignettes/datatable-faq.pdf" target="_blank">FAQs </a>of the data.table package.
Quirky and useful: when subsetting rows you can also use the symbol <code>.N</code> in the DT[&hellip;] command, which is the number of rows or the last row. You can use it for selecting the last row or an offset from it.
<pre><code>
DT[.N-1]
</code></pre>
<pre><code>
#Returns the penultimate row of DT
Year Month DayofMonth DayOfWeek DepTime ArrTime UniqueCarrier FlightNum TailNum ActualElapsedTime AirTime
2011 12 6 2 656 812 WN 621 N727SW 76 64
ArrDelay DepDelay Origin Dest Distance TaxiIn TaxiOut Cancelled CancellationCode Diverted
-13 -4 HOU TUL 453 3 9 0 0
</code></pre>
<h2>The j part</h2>
The &lsquo;j&rsquo; part is used to select columns and do stuff with them. And stuff can really mean anything. All kinds of functions can be used, which is a strong point of the data.table package.
<pre><code>
DT[, mean(na.omit(ArrDelay))]
</code></pre>
<pre><code>
[1] 7.094334
</code></pre>
Notice that the &#39;i&#39; part is left blank, and the first thing in&nbsp;the brackets is&nbsp;a comma. This might seem counterintuitive at first. However, this simply means that we do not subset on any rows, so all rows are selected. In the &#39;j&#39; part, the average delay on arrival of all flights is calculated. It appears that the average plane of the hflights dataset had more than 7 minutes delay. Be prepared when catching your next flight!
When selecting several columns and doing stuff with them in the &#39;j&#39; part, you need to use the &lsquo;<code>.()</code>&rsquo; notation. This notation is actually just an alias to &lsquo;<code>list()</code>&rsquo;. It returns a data table, whereas not using &lsquo;<code>.()</code>&rsquo; only returns a vector, as shown above.
<pre><code>
DT[, .(mean(na.omit(DepDelay)), mean(na.omit(ArrDelay)))]
</code></pre>
<pre><code>
V1 V2
9.444951 7.094334
</code></pre>
Another useful feature which requires the &lsquo;<code>.()</code>&rsquo; notation allows you to rename columns inside the DT[&hellip;] command.
<pre><code>
DT[, .(Avg_ArrDelay =
mean(na.omit(ArrDelay)))]
</code></pre>
<pre><code>
Avg_ArrDelay
7.094334
</code></pre>
<pre><code>
DT[, .(Avg_DepDelay = mean(na.omit(DepDelay)),
avg_ArrDelay = mean(na.omit(ArrDelay)))]
</code></pre>
<pre><code>
Avg_DepDelay Avg_ArrDelay
9.444951 7.094334
</code></pre>
Of course, new column names are not obligatory.
Combining the above about &lsquo;i&rsquo; and &lsquo;j&rsquo; gives:
<pre><code>
DT[UniqueCarrier==&amp;quot;AA&amp;quot;, .(Avg_DepDelay =
mean(na.omit(DepDelay)),
Avg_ArrDelay = mean(na.omit(ArrDelay)),
plot(DepTime,DepDelay,ylim=c(-15,200)),
abline(h=0))]
</code></pre>
<pre><code>
Avg_DepDelay Avg_ArrDelay V3 V4
6.390144 0.8917558 NULL NULL
</code></pre>
<a href="http://blog.datacamp.com/wp-content/uploads/2014/10/I-J-BY-1.png"><img alt="Learn the data.table package with DataCamp" class="aligncenter wp-image-395 size-medium" height="300" src="https://s3.amazonaws.com/assets.datacamp.com/blog_assets/data.table+Tutorial/I-J-BY-1-300x300.png" width="300" /></a>
Here we took DT, selected all rows where the carrier was AA in the &#39;i&#39; part, calculated the average delay on departure and on arrival, and plotted the time of departure against the delay on departure in the &#39;j&#39; part.
To recap, the &lsquo;j&rsquo; part is used to do calculations on columns specified in that part. As the columns of a data table are seen as variables, and the parts of &lsquo;j&rsquo; are evaluated as expressions, virtually anything can be done in the &lsquo;j&rsquo; part. This significantly shortens your programming time.
The by part
The final section&nbsp;of this data.table R tutorial&nbsp;focuses on the&nbsp;&lsquo;by&rsquo; part. The &lsquo;by&rsquo; part&nbsp;is used when we want to calculate the &lsquo;j&rsquo; part grouped by a specific variable (or a manipulation of that variable). You will see that the &lsquo;j&rsquo; expression is repeated for each &#39;by&#39; group. It is simple to use: you just specify the column you want to group by in the &#39;by&#39; argument.
<pre><code>
DT[,mean(na.omit(DepDelay)),by=Origin]
</code></pre>
<pre><code>
Origin V1
IAH 8.436951
HOU 12.837873
</code></pre>
Here, we calculated the average delay before departure, but grouped by&nbsp;where the plane is coming from. &nbsp;It seems that flights departing from HOU have a larger average delay than those leaving from IAH.
Just as with the &#39;j&#39; part, you can do a lot of stuff in the &#39;by&#39; part. Functions can be used in the &#39;by&#39; part so that results of the operations done in the &#39;j&#39; part are grouped by something we specified in the DT[&hellip;] command. Using functions inside <code>DT[&hellip;]</code> makes that one line very powerful. Likewise, the &lsquo;<code>.()</code>&rsquo; notation needs to be used when using several columns in the &#39;by&#39; part.
<pre><code>
DT[,.(Avg_DepDelay_byWeekdays = mean(na.omit(DepDelay))), by=.(Origin,Weekdays = DayOfWeek&amp;lt;6)]
</code></pre>
<pre><code>
Origin Weekdays Avg_DepDelay_byWeekdays
IAH FALSE 8.286543
IAH TRUE 8.492484
HOU FALSE 10.965384
HOU TRUE 13.433994
</code></pre>
Here, the average delay before departure of all planes (no subsetting in the &#39;i&#39; part, so all rows are selected) was calculated first, and grouped secondly, first by origin of the plane and then by weekday. Weekdays is <code>False </code>in the weekends. It appears that the average delay before departure was larger when the plane left from HOU than from IAH, and surprisingly the delays were smaller in the weekends.
Putting it all together a typical <code>DT[i,j,by]</code> command gives:

<pre><code>
DT[UniqueCarrier==&amp;quot;DL&amp;quot;, .(Avg_DepDelay =
mean(na.omit(DepDelay)),
Avg_ArrDelay = mean(na.omit(ArrDelay)),
Compensation = mean(na.omit(ArrDelay - DepDelay))), by = .(Origin, Weekdays = DayOfWeek&amp;amp;lt;6)]
</code></pre>
<pre><code>
Origin Weekdays Avg_DepDelay Avg_ArrDelay Compensation
IAH FALSE 8.979730 4.116751 -4.825719
HOU FALSE 7.120000 2.656566 -4.555556
IAH TRUE 9.270948 6.281941 -2.836609
HOU TRUE 11.631387 10.406593 -1.278388
</code></pre>

Here the subset of planes flewn by Delta Air Lines (selected in &#39;i&#39;) was grouped by their origin and by Weekdays (in &#39;by&#39;). The time that was compensated in air was also calculated (in &#39;j&#39;). It appears that in the weekends, irrespective of the plane was coming from IAH or HOU, the time compensated while in air (thus by flying faster) is bigger.
There is much more to discover in the data table package, but this post illustrated the basic <code>DT[i,j,by]</code> command.&nbsp;The <a href="https://www.datacamp.com/courses/data-analysis-the-data-table-way/?utm_source=blog&amp;utm_medium=blog&amp;utm_content=data.table%20i-j-by%20blog&amp;utm_campaign=data.table%20i-j-by%20blog" target="_blank" title="data.table R tutorial ">DataCamp course</a> explains&nbsp;the whole data table package extensively. You can do the exercises at your own pace in your browser while getting hints&nbsp;and feedback, and review the videos and slides as much as you want. This interactive way of learning allows you&nbsp;to gain profound knowledge and practical experience with data tables. &nbsp;<a href="https://www.datacamp.com/courses/data-analysis-the-data-table-way/?utm_source=blog&amp;utm_medium=blog&amp;utm_content=data.table%20i-j-by%20blog&amp;utm_campaign=data.table%20i-j-by%20blog" target="_blank" title="data.table R tutorial">Try it &nbsp;for free</a>.
Hopefully you know understand thanks to this data.table R tutorial the fundamental syntax of data.table, and are you ready to experiment yourself. If you have questions concerning the data.table package, have a look <a href="http://stackoverflow.com/questions/tagged/data.table" target="_blank">here</a>. Matt and Arun are very active.&nbsp;One of the next blogposts on the data.table package will be more technical, zooming in on the wide possibilities with data tables. Stay tuned!

https://www.datacamp.com/community/tutorials/data-table-r-tutorial

The free data.table R tutorial explains the basics and syntax of the data.table package for R. Master the data.table syntax now.

A data.table R tutorial by DataCamp:  intro to DT[i, j, by]

A data.table R Tutorial:  Intro to DT[i, j, by]

This post was written by the team behind <a href="https://www.datacamp.com/?utm_source=infographic&amp;utm_medium=blog%20&amp;utm_content=infographic%20how%20to%20become%20a%20data%20scientists&amp;utm_campaign=infographic%20how%20to%20become%20a%20data%20scientists" target="_blank">DataCamp</a>, the online interactive learning platform for data science. &nbsp;
After being dubbed &ldquo;sexiest job of the 21st Century&rdquo; by <a href="http://hbr.org/2012/10/data-scientist-the-sexiest-job-of-the-21st-century/ar/1" target="_blank">Harvard Business Review,</a> data scientists have stirred the interest of the general public. Many people are intrigued by this job, namely because the name has an interesting ring to it. But it is exactly the name that also raises a lot of questions. Because what is a data scientist and what do data scientists&nbsp;do exactly? Many of us who devote their lives to data science have frequently been confronted with questions like these.
The answers to these questions are mostly not as straightforward as you&nbsp;would expect: a short search on Google with the string of words &ldquo;How to become a data scientist&rdquo; shows that the concept has different meanings to different people. In addition, many articles indeed suggest various tools, courses and applications for people to become a data scientist, and with good reason: the options are unlimited.&nbsp;But let&rsquo;s face it, for someone that is not familiar with the field, this advice may sometimes seem like a jungle of information. What&rsquo;s more, they could work demotivating: the descriptions are sometimes fearfully long and the many details often hit the readers as an overwhelming avalanche.

<h2>DataCamp&rsquo;s Guide to Become&nbsp;a Data Scientist</h2>

 

With all this in mind, <a href="https://www.datacamp.com/?utm_source=infographic&amp;utm_medium=blog%20&amp;utm_content=infographic%20how%20to%20become%20a%20data%20scientists&amp;utm_campaign=infographic%20how%20to%20become%20a%20data%20scientists" target="_blank">DataCamp</a> decided to help those who can&rsquo;t see the forest&nbsp;for&nbsp;the trees:&nbsp;we designed a step-by-step infographic&nbsp;that clearly outlines how you&nbsp;can become a data scientist in 8 easy steps. &nbsp;This visual guide&nbsp;is&nbsp;meant for everyone that is interested in learning data science or for everyone that has already become a data scientist but wants some additional resources for further perfection. &nbsp;The infographic&nbsp;is called &ldquo;Become a data scientist in 8 easy steps&rdquo;. Have a look at it!


<a href="https://s3.amazonaws.com/assets.datacamp.com/blog_assets/Infographics/8-easy-steps-to-become-a-data-scientist.jpg" target="_blank"><img alt="How to become a data scientist" class="alignleft wp-image-386 " height="1015" src="https://s3.amazonaws.com/assets.datacamp.com/blog_assets/Infographics/How-to-become-a-data-scientist-223x1024.jpg" width="221" /></a> 

 
 
Source:&nbsp;<a href="https://www.datacamp.com/community/blog/">blog.datacamp.com</a>
If you are thinking about becoming a data scientist, do not be taken aback by the eight steps that are presented in the infographic. We would like to emphasize that becoming a data scientist takes time and personal investment, but that the&nbsp;journey is everything but dull! And don&#39;t forget, there are <a href="https://www.datacamp.com/courses/?utm_source=infographic&amp;utm_medium=blog%20&amp;utm_content=infographic%20how%20to%20become%20a%20data%20scientists&amp;utm_campaign=infographic%20how%20to%20become%20a%20data%20scientists" target="_blank">plenty of courses</a> available to set you on the right way.
If you are already a data scientist, drop us a line at <a href="mailto:%20info@datacamp.com">info@datacamp.com</a> if you think of other steps that you have undertaken in your professional journey.
 
Feel free to share!

https://www.datacamp.com/community/tutorials/how-to-become-a-data-scientist

This DataCamp infographic outlines how you can become a data scientist in 8 steps. This guide is for everyone interested in data science.

Become a Data Scientist in 8 Steps: The Infographic

Become A Data Scientist in 8 Steps: Infographic

<html>
 <body>
 Especially for useR! 2014 we created a poster on <a title="rdocumentation " href="http://www.rdocumentation.org/" target="_blank">Rdocumentation.org</a>, our R documentation aggregator that lets you search R documentation and R manuals from CRAN, Bioconductor and GitHub. We received a lot of positive and valuable feedback on it, so we decided to share it again via our blog.
 
 <a title=" R documentation and R manuals" href="https://s3.amazonaws.com/assets.rdocumentation.org/Rdocumentation+poster/Rdocs_poster.pdf" target="_blank"><img class="aligncenter wp-image-349" src="https://s3.amazonaws.com/assets.datacamp.com/blog_assets/Infographics/Screenshot-2014-06-28-16.36.26.png" alt=" R documentation and R manuals" width="196" height="270"></a>
 
 The <a href="https://s3.amazonaws.com/assets.rdocumentation.org/Rdocumentation+poster/Rdocs_poster.pdf" target="_blank">Rdocumentation poster</a> answers questions on R documentation and R manuals such as:
 <ul>
 <li>What are the 10 most downloaded R packages of all time?</li>
 <li>Who maintains the most packages?</li>
 <li>What are the most popular task views, and how did this ranking change over time?</li>
 <li>How popular is GitHub in the R community?</li>
 <li>...</li>
 </ul>
 
 So if you did not make it to the useR! 2014 conference, or in case you could not attend the Wednesday evening poster session, you hereby have another chance to have a detailed look at the poster. Feedback and questions can be sent to <a href="mailto:team@datacamp.com">team@datacamp.com</a>.
 Feel free to share it with your network!
 On <a href="http://www.rdocumentation.org/" target="_blank">Rdocumentation.org</a>: Rdocumentation is a web application that helps you to easily find and browse the documentation of R documentation and R manuals on CRAN, Bioconductor and GitHub. It enables you to instantly search for functions and use advanced search on the documentation of all R packages.
 </body>
</html>

https://www.datacamp.com/community/tutorials/package-rankings-task-view-rankings-and-much-more-the-rdocumentation-poster

Rdocumentation.org is the only R documentation and R manuals aggregator that lets you search everything from CRAN, Bioconductor and GitHub.

 R Documentation and R Manuals Rankings: the RDocumentation Poster

The RDocumentation Poster

As you all know, the decimal mark is a symbol used to separate the integer part from the fractional part of a number written in decimal form.&nbsp;Since I was born and raised in Continental Europe, I am quite fond of using the comma sign to indicate a decimal point. I&#39;ve grown up with it, encountered this comma in both my literary and numerical escapades, and still shudder when thinking of its dual role in long divisions. 
However, using the&nbsp;comma sign&nbsp;as a decimal mark has two consequences:

<ul>
	<li>Since the&nbsp;comma &quot;,&quot; sign is already taken to mark the radix point,&nbsp;I&#39;m obligated to use the&nbsp;dot &quot;.&quot; sign to separate the thousands.</li>
	<li>As the&nbsp;comma &quot;,&quot; sign is only used by 24% of the world&#39;s population, 76% of the people in the world are creating documents, writing texts, typing in working in spreadsheets, etc. that contain numbers with a fractional part that looks different from mine.</li>
</ul>
 
While the first consequence is the inevitable sacrifice one must make for having the privilege to use the comma for fractional numbers, the second is a much harder obstacle to deal with in the real world of professional number crunchers. More than necessary, I find myself struggling with sheets and documents that use&nbsp;the dot &quot;.&quot; sign to mark the radix point instead of my beloved comma. Why that often? <a href="http://en.wikipedia.org/wiki/Decimal_mark" target="_blank">Based on Wikipedia</a>,&nbsp;roughly 60% of the world&#39;s populations uses the dot &quot;.&quot; sign to mark the radix point. And when looking for these numbers, I learnt that there are even more dissidents. In the Arab world they use the Arabic decimal separator for Eastern Arabic numerals, in Persian the decimal mark is called momayyez, and in English Braille the decimal mark even has its own sign&hellip; So it gets difficult to know whether to user a decimal comma or decimal point. 
Since&nbsp;politely asking these people to change their disrupting behavior will most likely be hopeless, and finger pointing is only a prerogative of the majority, I was forced to make myself a little tool to guide me through of what at the beginning looked like an&nbsp;insurmountable problem. Using the data I found on Wikipedia and with the help of the <a href="http://www.rdocumentation.org/packages/googleVis" target="_blank">googleVis package</a>, I created a world map that indicates the decimal separator that is used in each country. That way, depending on the origin of the sheet I receive, I always know whether to use a&nbsp;decimal comma or decimal point. 
 
<iframe src="https://assets.datacamp.com/blog_assets/the_map.html" width="100%" height="300"></iframe>
 
The map (<a href="http://assets.datacamp.com/blog_assets/the_map.html" target="_blank">full-size</a>) is not yet complete (mainly in Africa there are some blank spots left). So for those that are aware of the prevailing decimal mark culture in these regions, just let me know in the comment section and I&#39;ll make sure to update them in case of a&nbsp;comma, or to proselytize them otherwise ;-). You can find the original code <a href="http://www.r-fiddle.org/#/fiddle?id=0KUiZpBJ" target="_blank">here</a>.

https://www.datacamp.com/community/tutorials/decimal-comma-or-decimal-point-a-googlevis-visualization

Decimal comma or decimal point? A handy tool to indicate the decimal separator used in each country so you never make a mistake in your sheet

Decimal Comma or Decimal Point? A googleVis Visualization

Decimal Comma or Point? A googleVis visualization

A feature that all programming communities have in common is the numerous debates about why their programming language of choice is better, more advanced, faster, holier etc. In today&#39;s data science community, it seems as if these discussions are omnipresent with advocates of SAS, SPSS, R, Python, Julia, etc. battling and challenging each other on every online medium on the best statistical programming language. (side note: These &#39;data driven&#39; debates are often a good example of how you can prove anything with statistics.) 
 
While these debates are a good thing for the community and the programming language as a whole, they unfortunately also have a negative effect on those individuals that are just in the beginning of their data analytics career. Biased opinions on all sides of the table make it difficult for new data analysts to see the forest for the trees when choosing a statistical programming language.

 

<h2>An Infograph for Each Statistical Programming Language</h2>


Especially for this new group of data analysts (and future debaters), as well as for everyone else that is interested in learning data science or an additional statistical language, we created the infograph &#39;Statistical Language Wars&#39; that gives a basic comparison between statistical programming languages like SAS, R and SPSS to see how they stack up. This is to provide a more clear starting point. 
&nbsp;

(To dive in to learning R, try this <a href="https://www.datacamp.com/courses/free-introduction-to-r/">free introductory course</a>.)

<a href="http://datacamp.wpengine.com/wp-content/uploads/2014/05/infograph.png" target="_blank" title="statistical programming language"><img alt="statistical programming language" class="alignnone" height="5671" src="http://datacamp.wpengine.com/wp-content/uploads/2014/05/infograph.png" width="2048" /></a> 
Source: <a href="https://www.datacamp.com/community/blog/">blog.datacamp.com</a>
 
We&#39;ll make sure to regularly update this infograph based on the feedback you provide, and we will definitely consider to create some new infographs that focus more on other players such as Python and Julia.
Feel free to share!

https://www.datacamp.com/community/tutorials/statistical-language-wars-the-infograph

The infograph 'Statistical Language Wars' compares statistical programming language like SAS, R and SPSS to see how they stack up.

What Is the Best Statistical Programming Language? Infograph

To give this years April Fools&#39; day a more analytical touch, we decided last week do a little poll on internet cartoons. We asked our friends and colleagues to select their favourite Data Science Cartoons&nbsp;on the web, and organized a voting session to construct a top 5 list. (You can always share your own favourites in the comments.) 
We proudly present you the winners of the April Fools&#39; 2014 Data Science Cartoons&nbsp;awards:

<h2>Number One: The Cloud&nbsp;</h2>

<a href="https://s3.amazonaws.com/assets.datacamp.com/blog_assets/Cartoons/cloud-cartoon.jpg"><img alt="Data Science Cartoons" class="alignnone wp-image-160" height="766" src="https://s3.amazonaws.com/assets.datacamp.com/blog_assets/Cartoons/cloud-cartoon.jpg" width="825" /></a>

<h2>Number Two: A Study on Statistics</h2>

 
<a href="https://s3.amazonaws.com/assets.datacamp.com/blog_assets/Cartoons/statistics.jpg"><img alt="Data Science Cartoons" class="alignnone wp-image-159 size-full" height="284" src="https://s3.amazonaws.com/assets.datacamp.com/blog_assets/Cartoons/statistics.jpg" width="700" /></a>


<h2>Number Three: Pacman Statistics</h2>

<a href="https://s3.amazonaws.com/assets.datacamp.com/blog_assets/Cartoons/pacman.jpg" title="Data Science Cartoons"><img alt="pacman" class="alignnone size-full wp-image-161" height="493" src="https://s3.amazonaws.com/assets.datacamp.com/blog_assets/Cartoons/pacman.jpg" width="500" /></a>

 

<h2>Number Four: Dilbert One</h2>

<a href="http://datacamp.wpengine.com/wp-content/uploads/2014/03/dilbertone.jpg" title="Data Science Cartoons"><img alt="dilbertone" class="alignnone size-full wp-image-162" height="205" src="http://datacamp.wpengine.com/wp-content/uploads/2014/03/dilbertone.jpg" width="618" /></a>

 

<h2>Number Five: Haloween Statistics</h2>

<a href="http://datacamp.wpengine.com/wp-content/uploads/2014/03/haloween.jpg" title="Data Science Cartoons"><img alt="haloween" class="alignnone size-full wp-image-163" height="720" src="http://datacamp.wpengine.com/wp-content/uploads/2014/03/haloween.jpg" width="486" /></a>

<h2>Number Six: Dilbert Two</h2>

<a href="http://datacamp.wpengine.com/wp-content/uploads/2014/03/dilbertwo.gif" title="Data Science Cartoons"><img alt="dilbertwo" class="alignnone size-full wp-image-164" height="200" src="http://datacamp.wpengine.com/wp-content/uploads/2014/03/dilbertwo.gif" width="640" /></a>

 

<h2>Number Seven: XKCD Correlation</h2>


<a href="http://datacamp.wpengine.com/wp-content/uploads/2014/03/correlation.png" title="Data Science Cartoons"><img alt="correlation" class="alignnone size-full wp-image-165" height="185" src="http://datacamp.wpengine.com/wp-content/uploads/2014/03/correlation.png" width="459" /></a>

 

<h2>Disqualified for the competition, but still one of the most funny data science cartoons:</h2>

<a href="http://datacamp.wpengine.com/wp-content/uploads/2014/03/big-data.jpg" title="Data Science Cartoons"><img alt="big data" class="alignnone size-full wp-image-166" height="504" src="http://datacamp.wpengine.com/wp-content/uploads/2014/03/big-data.jpg" width="540" /></a>

https://www.datacamp.com/community/tutorials/data-science-cartoons

The 7 funniest data science cartoons on the web selected by you. A selection made by DataCamp to celebrate April's fools day.

Data Science Cartoons: a Top Seven List by DataCamp

Like every <a href="https://www.datacamp.com/?utm_source=r-bloggers%20stack%20overflow%20artikel&amp;utm_medium=Link&amp;utm_campaign=r-bloggers%20stack%20overflow%20artikel" title="R tutorials">start-up in the IT and data science sector</a>, we often find ourselves spending more time on Stack Overflow than on our own site. For those of you who are not familiar with it, Stack Overflow is like a Q&amp;A forum on steroids. It features questions and answers on a wide range of topics in programming, and it&#39;s dedicated to answering any and all of these questions. Thanks to its clever reputation system that is based on points and badges, chances are high that you will find a high-quality answer to your particular problem. Believe us, this will save you a lot of time thanks to these R posts on Stack Overflow!

Since we use Stack Overflow that often for&nbsp;<a href="https://www.datacamp.com/?utm_source=r-bloggers%20stack%20overflow%20artikel&amp;utm_medium=Link&amp;utm_campaign=r-bloggers%20stack%20overflow%20artikel">DataCamp</a>, we wanted to share our &#39;Top Five&#39; list of the most popular R questions:

<a href="http://stackoverflow.com/questions/2039904/what-statistics-should-a-programmer-or-computer-scientist-know" target="_blank" title="Position 5">Position 5 = What statistics should a programmer (or computer scientist) know? (262 votes)</a><a href="http://stackoverflow.com/questions/2039904/what-statistics-should-a-programmer-or-computer-scientist-know%20%20%20%20" title="Position 5">&nbsp; &nbsp; </a>&nbsp;

This question is targeted at programmers who want to understand how their programming efforts can benefit from a more statistical approach. Not only does it provide an overview of statistical techniques, part of the answers also focus on the statistical tools programmers can use in their day-to-day activities.

<a href="http://stackoverflow.com/questions/3505701/r-grouping-functions-sapply-vs-lapply-vs-apply-vs-tapply-vs-by-vs-aggrega%20%20%20%20" target="_blank">Position 4 = R Grouping functions:&nbsp;sapply vs lapply vs apply vs tapply vs by vs aggregate&nbsp;(272 votes)&nbsp; &nbsp; &nbsp;</a>

This is something almost every new R programmer struggles with in the beginning: &nbsp;how and when to use the functions in the apply family. If you are one of these, just check out this Stack Overflow post and it will be a lot clearer to you. Multiple individuals have responded to the question, and most of them provide very clear answers with some even including slide presentations.

<a href="http://stackoverflow.com/questions/1296646/how-to-sort-a-dataframe-by-columns-in-r" target="_blank">Position 3 = How to sort a&nbsp;data frame&nbsp;by column(s) in R (302 votes)</a><a href="http://stackoverflow.com/questions/1296646/how-to-sort-a-dataframe-by-columns-in-r">&nbsp; &nbsp;</a>&nbsp; 

Again, a very easy but highly relevant question (certainly for new R users switching from Excel). Based on an example, the questioner wants to know how he can sort his data frame by multiple columns. This is a standard task in R, but if you&#39;re not familiar with using functions, the barrier to entry might be high.&nbsp;(Spoiler alert: the order function will take you a long way)

<a href="http://stackoverflow.com/questions/12675147/how-can-we-make-xkcd-style-graphs-in-r" target="_blank">Position 2 = How can we make&nbsp;xkcd&nbsp;style graphs in R (307 votes)&nbsp; &nbsp; &nbsp;</a>

Close, but no cigar. This question on xkcd style graphs reached the second place in our top five list. As a start-up, we personally love&nbsp;xkcd style graphs since they have this arty-farty layer over them. &nbsp;They allow you to provide information in a very clear way, but their unique and fun style just increases the chances that your audience will pick them up. A must read for everyone!

<a href="http://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example%20" target="_blank">Position 1 = How to make a great R reproducible example (525 votes)&nbsp;&nbsp;</a>

Simply put: great question and great answers! Reproducible examples are fundamental for teaching, research, and even for asking questions on, for example, Stack Overflow. However, the creation of reproducible examples is not that easy, and requires a certain finesse. This post will guide you through the ins and outs of creating such reproducible examples, so make sure to check it out since it will definitely help you to better understand R in the long run.

<a href="http://stats.stackexchange.com/questions/423/what-is-your-favorite-data-analysis-cartoon" target="_blank">Bonus: What&#39;s your favorite data analysis cartoon&nbsp;</a>

For the not&nbsp;so&nbsp;serious moments...

https://www.datacamp.com/community/tutorials/the-stack-overflow-r-top-5

Stack Overflow is a useful resource for technical questions. At DataCamp we wanted to share our favourite R posts on Stack Overflow.

R posts on Stack Overflow | A Top 5 selection by DataCamp

Coursera &ndash; an online education startup &ndash; has rapidly expanded its curriculum of statistics and data analysis courses. Today, there are already 33 modules that are directly linked to the field, excluding the courses where statistics and data science are solely used as a supportive tool (e.g. finance). These courses make use of multiple statistical software packages like Python, MATLAB and of course R. This post lists all&nbsp;Coursera courses on R.
I decided to make a list of all Coursera courses on R. These courses use R as either their first choice, or as one of the many statistical software packages that are allowed to be used by students to perform the homework&#39;s assignment. Coursera does not publish all data on how many students enroll in their courses, but most (some?) courses reach well over a hundred thousand students each year. 
To have some kind of indication of their popularity, I list below all Coursera courses on R&nbsp;ranked by the number of facebook likes:
<div class="output_wrapper">
<table border="0" cellpadding="0" cellspacing="0" class="tblGenFixed" dir="ltr" id="tblMain">
	<tbody>
		<tr>
			<td class="hd">&nbsp;</td>
			<td class="s0">&nbsp;</td>
			<td class="s1" dir="ltr">Ranking </td>
			<td class="s1" dir="ltr">Courese title</td>
			<td class="s1" dir="ltr">Professor</td>
			<td class="s1" dir="ltr">University</td>
			<td class="s1" dir="ltr">Facebook likes</td>
			<td class="s1" dir="ltr">Tweets</td>
		</tr>
		<tr>
			<td class="hd">&nbsp;</td>
			<td class="s2">&nbsp;</td>
			<td class="s3">1</td>
			<td class="s4" dir="ltr"> 
			<a href="https://www.google.com/url?q=https://www.coursera.org/course/sna&amp;sa=D&amp;usg=ALhdy2_k55L4ZvsIyHBIUfg1bzVmJWvdsA" target="_blank">Social Network Analysis</a></td>
			<td class="s4" dir="ltr">Lada Adamic</td>
			<td class="s4" dir="ltr">University of Michigan</td>
			<td class="s3">12000</td>
			<td class="s3">3543</td>
		</tr>
		<tr>
			<td class="hd">&nbsp;</td>
			<td class="s2">&nbsp;</td>
			<td class="s3">2</td>
			<td class="s4" dir="ltr"> 
			<a href="https://www.google.com/url?q=https://www.coursera.org/course/stats1&amp;sa=D&amp;usg=ALhdy2-QZKHUVChM3cBnCHNn63Rws1XnJA" target="_blank">Statistics one</a></td>
			<td class="s4" dir="ltr">Andrew Conway</td>
			<td class="s4" dir="ltr">Princeton University</td>
			<td class="s3">9600</td>
			<td class="s3">1421</td>
		</tr>
		<tr>
			<td class="hd">&nbsp;</td>
			<td class="s2">&nbsp;</td>
			<td class="s3">3</td>
			<td class="s4" dir="ltr"> 
			<a href="https://www.google.com/url?q=https://www.coursera.org/course/compdata&amp;sa=D&amp;usg=ALhdy28uj7uUnZtgjAyZv1h8LaOjUPEzgQ" target="_blank">Computing for Data Analysis</a></td>
			<td class="s4" dir="ltr">Roger Peng</td>
			<td class="s4" dir="ltr">John Hopkins University</td>
			<td class="s3">8500</td>
			<td class="s3">1934</td>
		</tr>
		<tr>
			<td class="hd">&nbsp;</td>
			<td class="s2">&nbsp;</td>
			<td class="s3">4</td>
			<td class="s4" dir="ltr"> 
			<a href="https://www.google.com/url?q=https://www.coursera.org/course/dataanalysis&amp;sa=D&amp;usg=ALhdy29k_d_X1cQpjiWrjRvNLxOfuoJDlQ" target="_blank">Data Analysis</a></td>
			<td class="s4" dir="ltr">Jeff Leek</td>
			<td class="s4" dir="ltr">John Hopkins University</td>
			<td class="s3">5200</td>
			<td class="s3">1408</td>
		</tr>
		<tr>
			<td class="hd">&nbsp;</td>
			<td class="s2">&nbsp;</td>
			<td class="s3">5</td>
			<td class="s4" dir="ltr"> 
			<a href="https://www.google.com/url?q=https://www.coursera.org/course/datasci&amp;sa=D&amp;usg=ALhdy2_A13sIWeO7TKb3axBFwkdOgR6g_Q" target="_blank">Introduction to Data Science</a></td>
			<td class="s4" dir="ltr">Bill Howe</td>
			<td class="s4" dir="ltr">University of Washington</td>
			<td class="s3">2600</td>
			<td class="s3">1103</td>
		</tr>
		<tr>
			<td class="hd">&nbsp;</td>
			<td class="s2">&nbsp;</td>
			<td class="s3">6</td>
			<td class="s4" dir="ltr"> 
			<a href="https://www.google.com/url?q=https://www.coursera.org/course/compfinance&amp;sa=D&amp;usg=ALhdy2_qhaARJUDCnqAIkZk5Ndd83XZLxg" target="_blank">Introduction to Computational Finance and Financial Econometrics</a></td>
			<td class="s4" dir="ltr">Eric Zivot</td>
			<td class="s4" dir="ltr">University of Washington</td>
			<td class="s3">2100</td>
			<td class="s3">351</td>
		</tr>
		<tr>
			<td class="hd">&nbsp;</td>
			<td class="s2">&nbsp;</td>
			<td class="s3">7</td>
			<td class="s4" dir="ltr"> 
			<a href="https://www.google.com/url?q=https://www.coursera.org/course/biostats&amp;sa=D&amp;usg=ALhdy2-dTsARXHGQwoyzPJjYzcnJk_r_vw" target="_blank">Mathematical Biostatistics Boot Camp 1</a></td>
			<td class="s4" dir="ltr">Brian Caffo</td>
			<td class="s4" dir="ltr">John Hopkins University</td>
			<td class="s3">1400</td>
			<td class="s3">239</td>
		</tr>
		<tr>
			<td class="hd">&nbsp;</td>
			<td class="s2">&nbsp;</td>
			<td class="s3">8</td>
			<td class="s4" dir="ltr"> 
			<a href="https://www.google.com/url?q=https://www.coursera.org/course/introstats&amp;sa=D&amp;usg=ALhdy2_SyYz17uG-xCkuLzgfalYu3BKpjA" target="_blank">Statistics: Making Sense of Data</a></td>
			<td class="s4" dir="ltr">Alison Gibs &amp; Jeffey Rosenthal</td>
			<td class="s4" dir="ltr">University of Toronto</td>
			<td class="s3">1400</td>
			<td class="s3">243</td>
		</tr>
		<tr>
			<td class="hd">&nbsp;</td>
			<td class="s2">&nbsp;</td>
			<td class="s3">9</td>
			<td class="s4" dir="ltr"> 
			<a href="https://www.google.com/url?q=https://www.coursera.org/course/assetpricing&amp;sa=D&amp;usg=ALhdy28-3VBuJvxdWNGcDqiU_flIWTTQqA" target="_blank">Asset Pricing</a></td>
			<td class="s4" dir="ltr">John H. Cochrane</td>
			<td class="s4" dir="ltr">University of Chicago Booth</td>
			<td class="s3">855</td>
			<td class="s3">102</td>
		</tr>
		<tr>
			<td class="hd">&nbsp;</td>
			<td class="s2">&nbsp;</td>
			<td class="s3">10</td>
			<td class="s4" dir="ltr"> 
			<a href="https://www.google.com/url?q=https://www.coursera.org/course/mathematicalmethods&amp;sa=D&amp;usg=ALhdy2-oeGj2ERjDSj9TJy-Q6ZqLjv6Haw" target="_blank">Mathematical Methods for Quantitative Finance</a></td>
			<td class="s4" dir="ltr">Kjell Konis</td>
			<td class="s4" dir="ltr">University of Washington</td>
			<td class="s3">635</td>
			<td class="s3">92</td>
		</tr>
		<tr>
			<td class="hd">&nbsp;</td>
			<td class="s2">&nbsp;</td>
			<td class="s3">11</td>
			<td class="s4" dir="ltr"> 
			<a href="https://www.google.com/url?q=https://www.coursera.org/course/casebasedbiostat&amp;sa=D&amp;usg=ALhdy2_X0G2zq4qLIorOLZj4fGQV4HIGEQ" target="_blank">Case-Based Introduction to Biostatistics</a></td>
			<td class="s4" dir="ltr">Scott L. Zeger</td>
			<td class="s4" dir="ltr">John Hopkins University</td>
			<td class="s3">424</td>
			<td class="s3">110</td>
		</tr>
		<tr>
			<td class="hd">&nbsp;</td>
			<td class="s2">&nbsp;</td>
			<td class="s3">12</td>
			<td class="s4" dir="ltr"> 
			<a href="https://www.google.com/url?q=https://www.coursera.org/course/fe2&amp;sa=D&amp;usg=ALhdy2-tSNaR93SYtcFExLE_UwERrQs36g" target="_blank">Financial Engineering 2</a></td>
			<td class="s4" dir="ltr">Martin Haugh &amp; Garud Iyengar</td>
			<td class="s4" dir="ltr">Coumbia University</td>
			<td class="s3">109</td>
			<td class="s3">13</td>
		</tr>
		<tr>
			<td class="hd">&nbsp;</td>
			<td class="s2">&nbsp;</td>
			<td class="s3">13</td>
			<td class="s4" dir="ltr"> 
			<a href="https://www.google.com/url?q=https://www.coursera.org/course/statistics&amp;sa=D&amp;usg=ALhdy28oxuO4zS2cMSpQH4DkkUWuTUGitA" target="_blank">Data Analysis and statistical inference</a></td>
			<td class="s4" dir="ltr">Mine &Ccedil;etinkaya-Rundel</td>
			<td class="s4" dir="ltr">Duke University</td>
			<td class="s3">80</td>
			<td class="s3">18</td>
		</tr>
		<tr>
			<td class="hd">&nbsp;</td>
			<td class="s2">&nbsp;</td>
			<td class="s3">14</td>
			<td class="s4" dir="ltr"> 
			<a href="https://www.google.com/url?q=https://www.coursera.org/course/datan&amp;sa=D&amp;usg=ALhdy2-JO-PCsS-HqksXgXmm5nFi0vCeYg" target="_blank">Core Concepts in Data Analysis</a></td>
			<td class="s4" dir="ltr">Boris Mirkin</td>
			<td class="s4" dir="ltr">Higher School of Economics</td>
			<td class="s3">77</td>
			<td class="s3">15</td>
		</tr>
		<tr>
			<td class="hd">&nbsp;</td>
			<td class="s2">&nbsp;</td>
			<td class="s3">15</td>
			<td class="s4" dir="ltr"> 
			<a href="https://www.google.com/url?q=https://www.coursera.org/course/biostats2&amp;sa=D&amp;usg=ALhdy2_NCtQtllgFB5fkZpP18GtRPaR7DQ" target="_blank">Mathematical Biostatistics Boot Camp 2</a></td>
			<td class="s4" dir="ltr">Brian Caffo</td>
			<td class="s4" dir="ltr">John Hopkins University</td>
			<td class="s3">60</td>
			<td class="s3">21</td>
		</tr>
	</tbody>
</table>
</div>
Given the unwillingness of Coursera&#39;s search function, I had to manually draft the list above. Therefore, it is possible that I overlooked some of the courses. Feel free to mention them in the comment section, and I will make sure to update the list. In case you are interested in taking (or teaching) interactive data analysis courses, make sure to have a look at our own educational startup <a href="https://www.datacamp.com/" title="R tutorials">DataMind</a>.
While I expect that most of you are familiar with Coursera, for those who don&#39;t, this is a quick summary: Coursera is one of the leading providers of Massive Open Online Courses (MOOCs). Today they have more than 100+ institutional partners offering 500+ courses to over 5 million students worldwide. So despite <a href="http://moreorlessbunk.wordpress.com/2013/09/06/why-coursera-and-udacity-are-the-worst-things-that-ever-happened-to-moocs/" target="_blank">being criticized by some</a>, it is becoming more and more clear that they are here to stay.

https://www.datacamp.com/community/tutorials/complete-list-of-coursera-courses-using-r-ranked-by-popularity

A list of all Coursera courses on R. These courses use R as their first choice, or as one of the statistical software packages for homework.

community

datacamp

Tutorials