Intro to Data Science Midterm

COMPSCI 181

Computer

A programmable machine designed to follow instructions

Program

Instructions in computer memory to make it do something

Programmer

A person who writes instructions (programs) to make a computer perform a task

Central Processing Unit (CPU)

The main hardware component responsible for executing instructions and performing calculations

Main Memory

The primary storage area in a computer where data and instructions are stored for immediate access

Secondary Memory / Storage

Non-volatile storage devices that retain data even when the computer is turned off

Input Devices

Devices that send information to the computer from outside, such as a keyboard or mouse

Output Devices

Devices that display or present information from the computer to the user, such as a monitor or printer

Random Access Memory (RAM)

Another term for main memory, which is volatile and erased when the program terminates or the computer is turned off

Byte

A unit of memory that consists of 8 consecutive bits and has its own unique address

Address

A unique number that identifies each byte in the main memory

Secondary Storage

Non-volatile storage media, such as hard drives or flash drives, that retain data when the program is not running or the computer is turned off

Machine Language

The language that the computer understands, consisting of binary numbers (0s and 1s)

Programming Languages

Languages used by programmers to write instructions for the computer to execute

Algorithm

A set of well-defined steps that a program follows to perform a task

Low-level Language

A programming language that communicates directly with the computer hardware and is often written in binary machine code

Data Science

A blend of various tools, algorithms, and machine learning principles with the goal to discover hidden patterns from raw data

Primarily used to make decisions and predictions making use of casual and prescriptive analytics, and machine learning

Cons of Business Intelligence (BI) tools

Most data used is unstructured or semi-structured and tools are not capable of processing said data

R language

Software environment for statistical computing and graphics

Supported by the R foundation

←

Assignment operator

#

Comments

Variable Name

These must start with a letter or a dot, and can only contain:

• Letters

• Numbers

• Underlines

• Dots

Data Type

These are what classify the different values present in R

• Numeric

• Logical

• Characters

as.integer( )

This is used to create or convert a value into the ‘integer’ type

print( )

The explicit function used to display a variable

paste( )

The concatenates strings and characters

Example:

print(paste(“My name is”, name, “.”))

^

Exponentation

*

Multiplication

/

Division

%%

Modulo

5 %% 2 = 1

%/%

Integer division

5 %/% 2 = 2

install.package

Imports additional features from the library into the R environment

Vectors

One-dimensional arrays that can hold:

• Numeric data

• Character data

• Logical data

c( )

Function used to concatenate and form a vector

Elements

Components of a vector or array

length( )

The number of members in a vector or array

seq( )

Creating a general sequence vector

rep( )

Creating a vector of replicated elements

:

This operator generates regular sequences

sum( )

Returns the total of all elements in the vector

prod( )

Returns the product of all elements in the vector

mean( )

Returns the sum of the vector divided by the length of the vector

sort( )

This can be used to sort vectors

Defaults to ascending order unless decreasing = TRUE is set

Value Coercion

When one value type is converted to another to better maintain the same primitive data type throughout the entire vector

is.element( )

This function is used to test if a vector contains a given element

unique( )

Finds the unique elements of a vector

rev( )

Reverses the elements

which.max/min( )

This shows the index of the max or min

fivenum( )

Shows the five-number summary of a set

Recycling Rule

If there are two vectors of unequal length, the shorter will be reused in order to match the longer vector

Negative Index

This will strip the member whose position has the same absolute value as this

sample( )

This creates a random set of data

Boolean Expression

This is an expression that evaluates to either TRUE or FALSE

!

Logical NOT operator

&

Logical AND operator

|

Logical OR operator

Selection by Comparison

This is when a comparison operator is used on an entire vector at once. The results are then displayed in order

plot( )

This creates a basic plot of a graph that is ready to be tweaked and altered

Plot Types

• ‘p’ for points (default value)

• ‘l’ for lines

• ‘b’ for both points and lines

• ‘o’ for both overplotted

• ‘h’ for histogram

• ‘s’ for stair steps

• ‘n’ for no plotting

names( )

How we assign names to vector members

TableName$Column

Accessing a column in an R function

Matrix

A two dimensional generalization of a vector

matrix( )

Function to create a matrix

• data =

• nrow =

• ncol =

• byrow =

x[y, z]

Accessing row y and column z

col/rownames( )

names( ) ← name_vector

Multiplying Matrices

We can only do this if the number of columns in the first matrix is the same as the number of rows in the second column

2 × 3 × 3 × 4

%*%

Matrix multiplication operator

c/rbind( )

These functions both create matrices by combining several factors of the same length

dim( )

This function returns the dimension of the matrix

t( )

The transpose of a matrix

det( )

Determinate of the matrix

solve( )

Inverse of the matrix

apply( )

Apply a function to the margins of a matrix

Traspose

This is a matrix whose rows are the columns of the original

diag( )

Used to create the identity matrix for a desired dimension

Deconstruction

When we apply the c( ) function to a matrix to combine all the vectors into one

Matrix[, -x]

Deleting a matrix column

col/rowSums( )

Used to calculate the totals for each column/row of a matrix

as.matrix( )

This function converts a data.table into a character matrix

Uses one of the columns in the table as the row names

data.matrix( )

Converts a data.table into a numeric matrix

round( )

This function rounds off values in the first argument to the specified number of places

barplot( )

Used to create a barplot

Can be used with either a vector or matrix

names.arg

Argument used when plotting that plots names of each bar/point

horiz = TRUE

This changes the orientation of a barplot to be horizontal

text(x, y, label )

This adds labels to a data plot

Specify values for the x variable, the y variable and the labels

Values placed on a plot

We can display the data’s values near data points/bars with a custom code:

text(b, matrix$column + 5, labels = as.character(matrix$column), col = ‘red’)

legend( )

This adds a legend to an R plot

Placed with locations such as ‘topleft’

col = heat.colors(x)

Creates x amount of the heat color pattern in plots

col = cm.colors(x)

Creates x amount of the cm color pattern in plots

col = topo.colors(x)

Creates x amount of the topo color pattern in plots

beside = TRUE

This argument can be used to create clustered bar plots

pairs( )

This can be used to create a matrix of scatter plots

Used to compare multiple values of a matrix at once

Point Shapes

lower.panel = NULL

This can be used on the upper or lower half of a scatterplot matrix

Used to only view half of the usual plots

apply(x, Margin, fun )

This allows us to make entry-by-entry changes to data frame and matrices

• x is a matrix

• Margin is:

  • 1 for rows

  • 2 for columns

  • fun is the function to be applied

cex

Numeric values indicating point size

col

Color name for points

Factors

Variables that take categorial variables and stores them in levels

Categorial variable

Variables that take values based on labels or names

Continuous variable

These can take any values