R For Data Science

What Is Data Science?

• Data Science is a discipline that uses statistics, data analysis, machine learning and their related methods to understand and analyze actual phenomena.

• It employs techniques and theories of mathematics, statistics, information science, and computer science, in particular the subdomains of machine learning, classification, cluster analysis, uncertainty quantification, computational science, data mining, databases, and visualization.

According to Wikipedia:

• Data Analysis is a descriptive process of inspecting, cleansing, transforming, and modeling data with the goal of discovering useful information, suggesting conclusions, and supporting decision-making.

• Data Mining is a particular data analysis technique for discovering patterns in large data sets. It focuses on modeling and knowledge discovery for predictive rather than purely descriptive purposes.

• Data Modeling is a process used to define and analyze the data requirements needed to support business processes within information systems in organizations.

• Business Intelligence comprises the strategies and technologies used by enterprises for the data analysis of business information to provide historical, current and predictive views of business operations.

R is used in two associated disciplines, which are often confused as also being Data Science:

• Machine Learning (ML) gives computer systems the ability to progressively improve performance of a specific task with data, without being explicitly programmed, and

• Artificial Intelligence (AI), in contrast to the natural intelligence displayed by humans and other animals, is intelligence demonstrated by machines through Intelligent Agents that perceive their environment and take actions to maximize the chance of successfully achieving their designers’ intended goals.

Although they overlap, Data Science, Machine Learning, and Artificial Intelligence produce different outcomes:

• Data Science produces insights
• Machine Learning produces predictions
• Artificial Intelligence produces actions

What Is R?

R is a functional programming language designed to manipulate, display, and analyse statistical data.

R is a free, case-sensitive, interpreted language that uses statistical techniques to perform Data Analysis, Data Mining, Data Modeling, and Machine Learning, and to create Artifical Intelligence and Business Intelligence.

• R is a functional programming language, not procedural.
• It is vector-based and highly-optimised to speedily and efficiently manipulate and display all types of numeric and graphical data.
• R holds data in memory for speedy access, so it runs
  • better on fast processors with lots of memory for large datasets
  • but can be run on an distributed cluster running Apache Hadoop, either
  • on a local network or on a Hadoop cloud service.
• R was developed from S, which was created in 1976 by John Chambers of Bell Labs.
• It was written by Ross Ihaka and Robert Gentleman at the University of Auckland, New Zealand.
• The name R comes partly from the names of the two authors, and partly as a play on S.
• R was conceived in 1992 with an initial version released in 1995, and a stable beta version in 2000.
• While there are some important differences, most S code can run unaltered in R.
• R is currently being actively developed by the R Development Core Team, which includes John Chambers.
• Development of R is supported by the R Project for Statistical Computing.
• R is freely available under the GNU General Public License. Pre-compiled binary versions are provided for various operating systems. These are available on local CRAN mirrors.

The source code for the R software environment is written primarily in C, Fortran, and R, highly-optimised for fast computation of large collections of statistical data.

The popularity of R has increased substantially in recent years. As of March 2018, R is in the top 20 of the TIOBE index, (Python was ranked #4 at that time).

Installing R

R must be installed on a computer as currently it is not included in any operating system. Precompiled versions are available for major systems including Windows, MacOS, and Linux. It may require compiling on some Linux distributions.

R is available from the Comprehensive R Archive Network (CRAN), via a multinational network of CRAN Mirrors. Canadian CRAN mirrors are hosted at:

• Simon Fraser University
• Manitoba Unix User Group
• University of Toronto
• Dalhousie University

Precompiled versions are available for:

• Download for Linux (SFU)
• Download for OS X (SFU)
• Download for Windows (SFU)

Command Line R

R can be run from the command line in a Terminal window:

• which R – find the location of the R executable
• R --version – show the currently installed version of R
• R – start an interactive R session

Version information is shown (as above) and the R command prompt > is available.

• quit() – quit an R session**

The R Console Interface

While R has a command line interface, a graphical interface is installed as an executable application R.app or R.exe for Windows, MacOS, and Linux. With minor platform differences, they are similar to this startup screen on OS X:

Above: The R Console startup screen in OS X 10.11.6 El Capitan

R commands can be run from the command prompt in R Console, similar to using the command line interface in a terminal program.

Access To R Help

Tooltip help appears on hovering the cursor over the icons in the Toolbar along the top of the Console window. Access to R Help files is invoked by entering help.start() at the command prompt. This will launch the R Help server on your computer and open a comprehensive Help page in your Web browser.

> help.start()
starting httpd help server ... done
If the browser launched by '/usr/bin/open' is already running, it is
    *not* restarted, and you must switch to its window.
Otherwise, be patient ...
> 

Above: The R Language Help page opens in your default Web browser

Data File Management

An R installation includes a treasure-trove of statistical data files. These can be seen using the Data Manager, available under the Packages & Data menu. Clicking on a dataset will display its documentation in the lower pane of the R Data Manager.

Above: The Data Manager lists the extensive statistical datasets included with R.

Package Management

R Console has a built-in Package Management system that can be invoked from the Packages & Data menu. The top pane of the R Package Manager window lists all installed packages. Clicking on a package brings up its Documentation in the lower window.

Above: The Package Manager window lists installed R packages and documentation.

R Packages