Visualization Design and User Interaction MultiDimensional Tree Network

Visualization Design and User Interaction: Multi-Dimensional, Tree, Network, Text

Multi-Dimensional Data Visualization

Multi-Dimensional data • Examples – A user in facebook • (name, gender, age, location, education, …) – A possible design for airplane wing • (length, width, thickness, weight, span, range)

Most Often Seen Design: Parallel Coordinate https: //syntagmatic. github. io/parallelcoordinates/

Pros and Cons • Pros – Can accommodated many dimensions – Can deal with both numerical and categorical data – Support the observation of relationship among dimensions (correlation) • Cons – Required trained eyes to read the chart – Location of dimensional axes affects the perception of the result.

Example • https: //bl. ocks. org/jasondavies/1341281

Other Visualization Techniques

Tree Visualization

What We Have Seen

Other Designs • For the same purposes – Better use the space – Focus + context issues

Hyperbolic trees After selection Original https: //www. youtube. com/ watch? v=pwpze 3 RF 55 o

Space Tree • http: //www. cs. umd. e du/hcil/spacetree/org chart. avi

Network Visualization

What We Have Seen

Social Network Visualization

http: //zhang. ist. psu. edu/demo/Social Net. Sense/Tree. Net. Viz. mov

Other Designs • Node. Trix – http: //www. youtube. com/watch? v=7 G 3 Mxy. Oc. HK Q

Network Visualization Tools • Some tools are available for you to directly use – Gephi • https: //www. youtube. com/watch? v=HJ 4 Hcq 3 YX 4 k – Pajek • https: //www. youtube. com/watch? v=ZPp. Ym. Op 4 Urw • These tools emphasize analysis more than visualization – Can do many calculations. – Interaction with networks is limited

Text Visualization

Different Levels in Analysis of Text • Corpora Corpus Document cluster Document Word • Corpora: All emails of DNC web server • Corpus: all emails by a person • Document cluster: messages on one specific issue • Document: a particular email message • Word: individual words

What to Display? • Raw text • Summarization of text • Measures: Words and frequency • Criteria: temporal evolution, group comparison, etc.

Raw Text • The content of documents, codes, books, etc. • Issues • Focus + Context • Type of text • Comparing text • Topics of documents

Focus+Context

Type of Information

Relationship Among Words, Sentences

Text. Arc

Comparing Text

When dealing with a large number of documents (text corpus), the details are not quite relevant, but topics are!

Topics in Computational Linguistic • Different from what we use • “What is the topic of your paper? ” • “Social network Visualization” • In linguistics: A bag of words • “What is the topic of this set of documents? ” • “network, visualization, edge, link, degree, radius, distance”

What Are Documents About?

Topics of A Corpus • Using natural linguistic processing (NLP) tools • Topic: bag-of-word with high frequencies of occurrence

Visualizing Topics

What Topics? • Large scale level: not about individual documents

How Topics Are Related? • How are those topic words related?

Trend of Topics

Topic Evolution in Image Processing

Topics Are Sensitive to Time!

Trends • Can be used to analyze • • Newspaper stories Emails Online forum discussions Product review Microblogs. .

Topics Can Be Narrow and Specific • Sentiments.

Problems of Topic Models for Visual Analytics • For the topics at the level of corpus, you need NLP tools to get topics first. • LDA (Latent Dirichlet Allocation) • Hard to map topics back to individual documents.

Possible to Compare Topics Among Corpora

But Hard to Check Topics and Documents

Topics At the Level of Document

Document Cards

Demo • https: //vimeo. com/6127783

Topic of a Document • Word clouds! • Also about the frequency of word occurrence.

Tools to Generate Word Clouds • http: //tagcrowd. com/ • http: //www. wordle. net/ • http: //www. wordclouds. com/ • Limitation • Size of word collection • Control of stop words

Word Clouds in Python • https: //github. com/amueller/word_cloud