# Vs hundreds or thousands

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: lot: Look at 2 dimensions at a 8me •  Units on each axis are diﬀerent! h\p://www.mzandee.net/~zandee/sta8s8ek/stat- online/chapter4/pearson.html 1 2/22/12 Gene Expression >3 Dimensions vs. “Really High Dimensions” •  Expression level for hundreds of genes •  For many trials (diﬀerent individuals/condi8ons) •  Discover correla8ons: genes that commonly work together or in opposi8on •  Obvious/intui8ve dimensions –  Posi8on, Orienta8on, Time, Temperature, Color, etc. vs. •  Hundreds or thousands of a\ributes –  May be ﬂoa8ng point values or binary values –  Stored as a “feature vectors” for each data point –  Nearest Neighbor calcula8ons become very expensive –  Visualiza8on seems impossible h\p://www.imbb.forth.gr/people/poirazi/researchEP.html h\p://www.bioss.ac.uk/~dirk/essays/GeneExpression/bayes_net.html Today’s Class •  •  •  •  •  •  •  •  Parallel Coordinates Readings for this Week Examples of High Dimensional Data Parallel Coordinates Data Clustering Principle Components Analysis (PCA) General Massive Data Visualiza8on Tips Next Week’s Readings Assignment 5 & Mid- Term Presenta8on h\p://eagereyes.org /techniques/parallel- coordinates Parallel Coordinates Designing Visualiza8ons using Parallel Coordinates How many dimensions (ver8cal axes)? In what order should the axes appear? Which direc8on should each axis run (up or down?) How many data points (lines)? How could color...
View Full Document

## This document was uploaded on 03/16/2014 for the course CSCI 4973 at Rensselaer Polytechnic Institute.

Ask a homework question - tutors are online