Quiz #1

the ubiquity of data opportunities

generation, collection, accessibility, risk management, and decision making

→ computers became very powerful
→ data networking is very fast
→ computer storage is very cheap
→ algorithms were developed to process large datasets quickly

comparative size of different volumes of storage

→ bit (b): the smallest unit of data; can represent a 0 or 1
→ byte (B): consists of 8 bits; it is the fundamental unit of storage in computers
→ kilobyte (KB): equal to 1,024 bytes; often used to describe the size of small text files or images
→ megabyte (MB): equal to 1,024 KBs or 1,048,576 bytes; commonly used to measure the size of documents, photos, and short videos
→ gigabyte (GB): equal to 1,024 MBs or 1,073,741,824 bytes; used for larger files such as high-resolution videos, software applications, and larger databases
→ terabyte (TB): equal to 1,024 GBs or 1,099,511,627,776 bytes; often used to measure the storage capacity of hard drives, servers, and cloud storage

structured data

data that resides in a pre-defined row/column format 
→ 20% of data

unstructured data

does not conform to a pre-defined row/column format
→ 80%
→ ex. emails, videos, images, etc.

4 V’s of Big Data

1. volume = amount

2. variety = type of data (structured or unstructured)

3. velocity = speed

4. veracity = quality, trustworthiness of data, lack of bias, noise, and abnormalities 

types of data analytics

→ descriptive
→ diagnostic
→ predictive
→ prescriptive

important business analytics applications

consumer analytics, financial chain, HR, risk, etc.

supervised learning

discovering patterns in the data that relate data attributes / variables with a target variable

unsupervised learning

the data has no target variable
→ we want to explore the data to find some intrinsic structures or similarities in them

types of data

→ numerical
→ categorical (nominal, ordinal)

descriptive data analytics

hindsight, low value and low difficulty foresight, high value and high difficulty
→ what happened?

diagnostic data analytics

insight, understanding, mid-value and difficulty
→ why did it happen?

predictive data analytics

insight/foresight, forecasting, mid-value and difficulty
→ what will happen?

prescriptive data analytics

foresight, high value and high difficulty
→ how can we make it happen?

two categories of supervised learning

1. classification

2. regression

classification (supervised learning)

learns a method for predicting the instance class from pre-labeled instances

regression (supervised learning)

an attempt to predict a continuous attribute

pre-attentive attributes

visual characteristics of objects or elements that the human brain can quickly and automatically perceive without conscious effort or focused attention

most used pre-attentive attributes

→ size
→ shape
→ color (hue)

primary colors

red, yellow, blue

secondary colors

purple, green, and orange

tertiary colors

red-orange, yellow-orange, yellow-green, blue-green, blue-violet, and red-violet

color schemas

→ categorical
→ sequential
→ diverging

categorical (color schemas)

contrasting colors for individual comparison

sequential (color schemas)

color is ordered from low/light to high/dark

diverging (color schemas)

two sequential colors with a neutral midpoint (odd #)

what does CVD stand for?

color vision deficiency (color blindness)
→ most prevalent form is red/green CVD

three types of color sensitive cones:

1. short - respond to short wave lengths; sensitive to blue colors

2. medium - respond to medium wave lengths; sensitive to green colors

3. long - respond to long wave lengths; more sensitive to red colors