Press "Enter" to skip to content

Category: R

Twitter Sentiment Analysis Using doc2vec

Published 2017-02-16 by Kevin Feasel

Sergey Bryl uses word2vec and doc2vec to perform Twitter sentiment analysis in R:

But doc2vec is a deep learning algorithm that draws context from phrases. It’s currently one of the best ways of sentiment classification for movie reviews. You can use the following method to analyze feedbacks, reviews, comments, and so on. And you can expect better results comparing to tweets analysis because they usually include lots of misspelling.

We’ll use tweets for this example because it’s pretty easy to get them via Twitter API. We only need to create an app on https://dev.twitter.com (My apps menu) and find an API Key, API secret, Access Token and Access Token Secret on Keys and Access Tokens menu tab.

Click through for more details, including code samples.

Comments closed

Galaxy Classification With SQL Server

Published 2017-02-16 by Kevin Feasel

David Smith points out a nice Microsoft demo for classifying galaxies using SQL Server:

The SQL Server Blog has since published a step-by-step tutorial on implementing the galaxy classifier in SQL Server (and the code is also available on GitHub). This updated version of the demo uses the new MicrosoftML package in Microsoft R Server 9, and specifically the rxNeuralNet function for deep neural networks. The tutorial recommends using the Azure NC class of virtual machines, to take advantage of the GPU-accelerated capabilities of the function, and provides details on using the SQL Server interfaces to train the neural netowrk and run predictions (classifications) on the image database. For the details, follow the link below.

If you’re going to get into SQL Server R Services at any level of seriousness, I highly recommend R Tools for Visual Studio, as it will make building those external stored procedure calls much easier.

Comments closed

RStudio Connect

Published 2017-02-13 by Kevin Feasel

Jen Underwood discusses RStudio Connect:

RStudio officially introduced the newest product in RStudio’s product lineup: RStudio Connect. RStudio Connect is a new publishing platform for R that allows analytics users to share Shiny applications, R Markdown reports, dashboards, plots, and more. This release adds an improved user experience for parameterized R Markdown reports, simple button-click publishing from the RStudio IDE, scheduled execution and distribution of reports, and more security policies include hybrid data connections. Essentially RStudio Connect eases enterprise deployment scenarios.

Between what Microsoft is doing with its old Revolution Analytics holdings and what RStudio is doing, this is a great time to be an enterprise R customer.

Comments closed

Scaling Up R

Published 2017-02-13 by Kevin Feasel

Ginger Grant explains how to use SQL Server R Services to take advantage of server resources instead of running from your local machine:

Microsoft’s R Server contains some specialized functions which are not part of the standard CRAN R installation. One of the ScaleR functions, RxInSqlServer will allow code to be processed on the server from the client. To make this work, you must have R Server and R Client installed. If you are doing a test on a local machine, you will need both R Client and R Server installed on that computer.

Click through for a script which walks you through the process.

Comments closed

Dplyr Tutorial

Published 2017-02-09 by Kevin Feasel

Deepanshu Bhalla has a nice dplyr tutorial:

What is dplyr?
dplyr is a powerful R-package to manipulate, clean and summarize unstructured data. In short, it makes data exploration and data manipulation easy and fast in R.
 
What’s special about dplyr?

The package “dplyr” comprises many functions that perform mostly used data manipulation operations such as applying filter, selecting specific columns, sorting data, adding or deleting columns and aggregating data. Another most important advantage of this package is that it’s very easy to learn and use dplyr functions. Also easy to recall these functions. For example, filter() is used to filter rows.

dplyr is a core package when it comes to data cleansing in R, and the more of it you can internalize, the faster you’ll move in that language.

Comments closed

Looping In R

Published 2017-02-08 by Kevin Feasel

Klodian Dhana explains how to build a for loop in R:

I used linear mixed effect model and therefore I loaded the lme4 library. The loop should work with other regression analysis (i.e. linear regression), if you modify it according to your regression model. If you don’t know which part to modify, leave a comment below and I will try to help.

As other loops, this call variables of interest one by one and for each of them extract and store the betas, standard error and p value. Remember, this code is specific for linear mixed effect models.

Read the whole thing.  It’s good to keep in mind, though, that set-based R operations tend to perform best, so save looping for cases in which you can’t build a set-based function.

Comments closed

Mapping Geospatial Data

Published 2017-02-08 by Kevin Feasel

The folks at Sharp Sight Labs have a great blog post on mapping geospatial data using R:

If you’ve learned the basics of data visualization in R (namely, ggplot2) and you’re interested in geospatial visualization, use this as a small, narrowly-defined exercize to practice some intermediate skills.

There are at least three things that you can learn and practice with this visualization:

  1. Learn about color: Part of what makes this visualization compelling are the colors. Notice that in the area surrounding the US, we’re not using pure black, but a dark grey. For the title, we’re not using white, but a medium grey. Also, notice that for the rivers, we’re not using “blue” but a very specific hexadecimal color. These are all deliberate choices. As an exercise, I highly recommend modifying the colors. Play around a bit and see how changing the colors changes the “feel” of the visualization.

  2. Learn to build visualizations in layers: I’ve emphasized this several times recently, but layering is an important principle of data visualization. Notice that we’re layering the river data over the USA country map. As an exercise, you could also layer in the state boundaries between the country map and the rivers. To do this, you can use map_data().

  3. Learn about ‘Spatial’ data: R has several classes for dealing with ‘geospatial’ data, such as ‘SpatialLines‘, ‘SpatialPoints‘, and others. Spatial data is a whole different animal, so you’ll have to learn its structure. This example will give you a little experience dealing with it.

I also like the iterative approach they discuss.  You’ll almost never get it right the first go-around, but one of the nice things about ggplot2 is that it’s designed to be iterative:  you layer your changes on, making it a bit easier to fiddle with them to get the visualization just right.

Comments closed

Articles On R And SQL Server

Published 2017-02-08 by Kevin Feasel

Tomas Kastrun links to a series of his published articles on working with SQL Server and R:

In past couple of months, I have prepared several articles on R and SQL Server that have been published on SQL Server Central.

The idea was, to have couple of articles covering the introduction to R, to basics on R Server, to some practical cases on R with SQL Server.

There’s a nice flow here, building up from the basics to practical marginal improvements.

Comments closed

Custom Power BI Shapes Using R

Published 2017-02-08 by Kevin Feasel

Koen Verbeeck uses R to create dynamically changing images in Power BI:

You can insert images into Power BI Desktop, but these are static images. If you want them to dynamically change, you need the Image Viewer custom visual. Unfortunately, it doesn’t support measures, only columns. Since we want dynamic changes, fixed column values are not going to work. Someone proposed a work around on the Power BI forums, but this only works if you have a fixed set of attributes you want to slice on (for example, 4 categories). I want a totally flexible solution (e.g. each month we have a couple of new weeks to slice on), so again, not possible.

The only solution I could think of that would still work: using R visuals.

Read on for the solution.

Comments closed

Using Sparklyr To Analyze Flight Data

Published 2017-02-07 by Kevin Feasel

Aki Ariga uses sparklyr on Apache Spark 2.0 to analyze flight data living in S3:

Using sparklyr enables you to analyze big data on Amazon S3 with R smoothly. You can build a Spark cluster easily with Cloudera Director. sparklyr makes Spark as a backend database of dplyr. You can create tidy data from huge messy data, plot complex maps from this big data the same way as small data, and build a predictive model from big data with MLlib. I believe sparklyr helps all R users perform exploratory data analysis faster and easier on large-scale data. Let’s try!

You can see the Rmarkdown of this analysis on RPubs. With RStudio, you can share Rmarkdown easily on RPubs.

Sparklyr is an exciting technology for distributed data analysis.

Comments closed