DSC 465 - Notes 2.docx - Module 1 Data Visualization How to...

This preview shows page 1 - 5 out of 22 pages.

Module 1: Data Visualization How to use visualization well? - Data specifics of what data are available - Audience who will be using this? - Message what do you want the audience to see? Types of Visualization 1) Exploratory a. Where data analysis begins – get an understanding of the data b. Discoveries arise from data patterns i. Difficult without graphical techniques ii. Visualizations draw out subtle patterns c. “impressive techniques” should not distract from the data d. Audience is me – make sure it makes sense, made quickly and do many at a time 2) Explanatory a. Draft … analyze … rewrite b. Clean and clear – tell a story, mind message and audience c. Go through many drafts of visualization Module 1: Mapping and Perception Variables 1) Nominal (discreate, qualitative, categorial) a. Ex. Variable with specific set of variables such as animals dog, cat, pig b. You can compare (equal, less than, greater than) 2) Ordinal a. It has an order to it b. You can compare (equal, less than, greater than) c. You can subtract d. No division e. Ex. Months Jan, Feb, … etc. i. You can write them as numbers, but they are not numbers 3) Numerical (quantitative, continuous, ratio scale) a. Ex. 1234557689 etc. b. Variable can have any number assigned to it c. You can compare, subtract, and divide Pre-attentive attributes - Visual cues that are visually apparent very quickly (~200ms) and do not require intentional processing to see
Reasons to use data visualization - Take advantage of human visual processing machinery for insight generation - Can get high information density in an image - Conveys patterns that may be difficult to describe Representation - Discrete mapping identity mapping - Continuous mapping Magnitude mapping Color Subtlety - Value o brightness, amount of light coming through or reflecting o Luminance (white to black) o Saturation - Color o Hue Gestalt Psychology - Our brains are wired for certain types of patterns and this affects how we see the world
- Human brain finds structure in visual stimuli 1) Law of proximity o Things that are close together, get paired together Vertical Lines and horizontal lines 2) Law of Similarity o Shades or shaped 3) Connectedness and Containment can Overrule Description of Mapping: Channel interference: Weber’s Law - Relative nature of our perception – Human perception works by percentage increase - Perception is not simply on absolute terms, but we perceive differences as a function of the magnitude of the original (i.e. we perceive percentage changes)
Steven’s Power Law - Stimulus = Intensity of actual stimulus ^ n (length) - Perception of stimuli is not linear - Shows relationship between actual and perceived value ** Major differences in the character of datasets can be invisible to summary statistics ** Module 2: Hierarchical Data Hierarchical Data - Has top, middle, and bottom sections like a “pyramid” where it has subdivisions and main categories - Usually represented in a table it contains the main category, sub category, sub sub category, etc.

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture