lecture 01: introduction
DESCRIPTION
Lecture 01: Introduction. September 7, 2010 COMP 150-12 Topics in Visual Analytics. COMP 150-12: Topics in Visual Analytics. Fall 2010 TR 12:00-1:15pm Halligan 111A. People. Instructor: Remco Chang [email protected] Office: Halligan E009 Hours: ??. TA: Samuel Li - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/1.jpg)
LECTURE 01:
INTRODUCTION TO VISUAL ANALYTICSJanuary 14, 2015
COMP 150-04Topics in Visual Analytics
![Page 2: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/2.jpg)
COMP 150-12: Topics in Visual Analytics
Fall 2014MW 6:00-7:15pmLecture: Halligan 111ALab: Halligan 118
![Page 3: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/3.jpg)
People
Instructor:R. Jordan [email protected]
Office Hours: W 7:15-8:15pLocation: TBD
Teaching Assistant: TBD
![Page 4: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/4.jpg)
People•3 Minute Biographies:
• Your (preferred) name• Your major / area of focus• Year (grad vs. undergrad)
- If grad, where did you do your undergrad?• Technical background
- Programming language(s) you know/like- Any experience in web design, visualization, HCI
•2 Questions:• What do you hope to get out of this course?• What’s one problem / curiosity / issue that sometimes keeps
you up at night?
![Page 5: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/5.jpg)
Outline
• A quick history lesson• Visual Analytics: a definition• Why Visual Analytics?• Building blocks: perception• Structure of this course• Takeaways
![Page 6: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/6.jpg)
(Incomplete) History of Visual Analytics: 1970s
- CAD/CAM, building cars, planes, chips- Starting to think about: 3D, animation, edu, medicine
![Page 7: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/7.jpg)
(Incomplete) History of Visual Analytics: 1980s
- Scientific visualization, physical phenomena- Starting to think about: photorealism, entertainment
![Page 8: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/8.jpg)
(Incomplete) History of Visual Analytics: 1990s
- Information visualization, storytelling- Starting to think about: online spaces, interaction
![Page 9: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/9.jpg)
(Incomplete) History of Visual Analytics: 2000s
- Coordination across multiple views, interaction- Starting to think about: sensemaking, provenance
![Page 10: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/10.jpg)
(Incomplete) History of Visual Analytics• Early 2000s: US is reacting to 9/11• 2003: Dept. of Homeland Security (DHS) est.• DHS Goals:
- Prevent terrorist attacks within the US- Reduce US vulnerability to terrorism- Minimize damage / aid recovery from attacks that do
occur• 2005: DHS charters the National Visualization
and Analytics Center (NVAC) at PNNL
![Page 11: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/11.jpg)
(Incomplete) History of Visual Analytics
NVAC mission:Develop advanced information technologies to support the Homeland Security mission with data that is massive, complex, incomplete, and uncertain in scenarios requiring human judgment.
Challenges:• How do we support analytical reasoning under
complex, changing circumstances?• How do we make use of domain expertise, when
domain experts are not computer scientists?
New idea: (visualization) ∩ (analytics)
![Page 12: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/12.jpg)
Visualization (def.)
Creating visual representations
of data to reinforce human
cognition
![Page 13: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/13.jpg)
Analytics (def.)
Discovery and communication
of meaningful patterns in data
![Page 14: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/14.jpg)
Visual Analytics (def.)
“The science of analytical reasoning facilitated by interactive visual interfaces”1
1Thomas, James J., and Kristin A. Cook, eds. Illuminating the path: The research and development agenda for visual analytics. IEEE Computer Society Press, 2005.
![Page 15: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/15.jpg)
What is Visual Analytics?
Visualization plus…• data representation• interaction & analysis• dissemination & story telling• a scientific approach • (evaluation)
US Congress: “Visual analytics provides the last 12 inches between the masses of information and the human mind to make decisions.”
![Page 16: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/16.jpg)
Examples of Visual Analytics Systems(Financial Fraud)• Wire Fraud Detection
– With Bank of America– Hundreds of thousands
of transactions per day
• Global Terrorism– Application of the
investigative 5 W’s
• Bridge Maintenance – With US DOT– Exploring subjective
inspection reports
• Biomechanical Motion– Interactive motion
comparison methods
Chang, Remco, et al. "Scalable and interactive visual analysis of financial wire transactions for fraud detection." Information visualization 7.1 (2008): 63-76.
![Page 17: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/17.jpg)
Examples of Visual Analytics Systems(Analysis of Civil Unrest)• Wire Fraud Detection
– With Bank of America– Hundreds of thousands
of transactions per day
• Global Terrorism– Application of the
investigative 5 W’s
• Bridge Maintenance – With US DOT– Exploring subjective
inspection reports
• Biomechanical Motion– Interactive motion
comparison methods
Godwin, Alex, et al. "Visual analysis of entity relationships in the Global Terrorism Database." SPIE Defense and Security Symposium. International Society for Optics and Photonics, 2008.
![Page 18: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/18.jpg)
Examples of Visual Analytics Systems(Transportation Analysis)• Wire Fraud Detection
– With Bank of America– Hundreds of thousands
of transactions per day
• Global Terrorism– Application of the
investigative 5 W’s
• Bridge Maintenance – With US DOT– Exploring subjective
inspection reports
• Biomechanical Motion– Interactive motion
comparison methods
Wang, Xiaoyu, et al. "An interactive visual analytics system for bridge management." Computer Graphics Forum. Vol. 29. No. 3. Blackwell Publishing Ltd, 2010.
![Page 19: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/19.jpg)
Examples of Visual Analytics Systems(Biomechanical Motion)• Wire Fraud Detection
– With Bank of America– Hundreds of thousands
of transactions per day
• Global Terrorism– Application of the
investigative 5 W’s
• Bridge Maintenance – With US DOT– Exploring subjective
inspection reports
• Biomechanical Motion– Interactive motion
comparison methodsSpurlock, Scott, et al. "Combining automated and interactive visual analysis of biomechanical motion data." Advances in Visual Computing. Springer Berlin Heidelberg, 2010. 564-573.
![Page 20: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/20.jpg)
Other Examples of Visual Analytics Systems(Pandemic / Healthcare)
• Healthcare– Unreliable data sources– Spatiotemporal analysis
• Network Security– Large amounts of
transactional data
• Energy / Power Grid– Graph-based
visualization– Identifies failure points
in the system
• Multimedia Analysis– Text analysis– Image and video
analysis
Maciejewski, Ross, et al. "A visual analytics approach to understanding spatiotemporal hotspots." Visualization and Computer Graphics, IEEE Transactions on 16.2 (2010): 205-220.
![Page 21: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/21.jpg)
Other Examples of Visual Analytics Systems(Network Security)
• Healthcare– Unreliable data sources– Spatiotemporal analysis
• Network Security– Large amounts of
transactional data
• Energy / Power Grid– Graph-based
visualization– Identifies failure points
in the system
• Multimedia Analysis– Text analysis– Image and video
analysis
Interactive Wormhole Detection in Large Scale Wireless Networks, Weichao Wang, Aidong Lu, Proceedings of IEEE Symposium on Visual Analytics Science and Technology (VAST), pp.99-106, 2006.
![Page 22: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/22.jpg)
Other Examples of Visual Analytics Systems(Energy / Power Grid)
• Healthcare– Unreliable data sources– Spatiotemporal analysis
• Network Security– Large amounts of
transactional data
• Energy / Power Grid– Graph-based
visualization– Identifies failure points
in the system
• Multimedia Analysis– Text analysis– Image and video
analysis
Wong, Pak Chung, et al. "A novel visualization technique for electric power grid analytics." Visualization and Computer Graphics, IEEE Transactions on 15.3 (2009): 410-423.
![Page 23: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/23.jpg)
Other Examples of Visual Analytics Systems(Multimedia Analysis)
• Healthcare– Unreliable data sources– Spatiotemporal analysis
• Network Security– Large amounts of
transactional data
• Energy / Power Grid– Graph-based
visualization– Identifies failure points in
the system
• Multimedia Analysis– Text analysis– Image and video
analysisH. Luo, et al., ``Integrating Multi-Modal Content Analysis and Hyperbolic Visualization for Large-Scale News Videos Retrieval and Exploration.” Signal Processing: Image Communication, vol.23, no.8, pp.538-553, 2008.
![Page 24: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/24.jpg)
Why Visual Analytics?
• We are collecting and generating data faster than traditional methods can keep up
• This data is often complex, ambiguous, noisy Not only is the data unmanageably big, it requires
human interpretation and understanding Oh, great
• Major problem: humans don’t scale, either
![Page 25: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/25.jpg)
VA: Human + Machine Collaboration• Make use of what both humans and machines
bring to the table using visual interfaces as a medium
• Key considerations:- Traditional analytics (stats, machine learning) can
help make massive data tractable- We can use what we know about
cognition/perception to help analysts put data in context
![Page 26: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/26.jpg)
Perception: Preattentive Processing
• First impressions matter: <200ms of visual stimulation• Performed in parallel across the entire visual field• Detects basic features such as:
– Color– Intensity– Size– Density– Line termination– Intersection
• Facilitates several important tasks:– Target detection (presence or absence)– Boundary detection / grouping– Region tracking– Counting and estimation
– Curvature– Closure– Tilt– Light source direction– Flicker– Velocity
![Page 27: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/27.jpg)
Perception: Preattentive Processing
![Page 28: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/28.jpg)
Perception: Preattentive Processing
• What did you see?
![Page 29: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/29.jpg)
Perception: Preattentive Processing
• Whatever draws our eyes draws our attention• In visual analytics, we can use preattentive processing to
our advantage:- Alerts- Anomalies- Situational awareness
• However, this same • processing can also• be problematic:- Artifacts- Visual distractors- Change blindness
![Page 30: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/30.jpg)
Perception: Preattentive Processing
![Page 31: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/31.jpg)
Takeaways: Perception
• Not just pretty pictures!• There are compelling cognitive reasons why some
visualization techniques are helpful and others… not so much
• Low-level decisions about visual mappings can have a significant effect on overall performance- Analogous to design choices in algorithms, etc.- Need to understand the impact on efficiency and
accuracy/integrity- Manage the analyst’s cognitive burden
![Page 32: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/32.jpg)
What We’ll Cover in Lecture• Next Class: Mental and Visualization Models • Unit 1: Data Wrangling
- Data Collection and Cleaning- Analysis Tools- Data Modeling
• Unit 2: Visualization Techniques- Introduction to Visualization- Visual Mapping- Data Projections- Interaction- Storytelling with Visual Analytics
• Unit 3: Advanced Topics- Practical Challenges in Building VA Systems- Analytic Provenance- Evaluation Techniques- Open Research Topics
![Page 33: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/33.jpg)
Structure of This Course
• Disclaimer: this class is an experiment in constructionism (the idea that people learn most effectively when they’re building meaningful things)
• My focus as an instructor:- Expose you to some foundational principles and available tools- Ask questions that build your intuitions about identifying
problems where VA techniques can help- Most important: help you find opportunities to solve real
problems in areas YOU care about (and hopefully learn some cool stuff about VA along the way…)
![Page 34: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/34.jpg)
Guest Tutorials and Demonstrations
Maja Milosevjivic, MITLL / Smith College- Introduction to R
Lane Harrison, Tufts / UNCC- Crash Course in D3.js
Megan Monroe, IBM / UMD- EventFlow and Semantic Interaction
Rajmonda Caceres, MITLL / UIC- Data Projections
Diane Staheli, MITLL- Conducting a Needs Assessment
…and more TBD!
![Page 35: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/35.jpg)
Assignments and Grading• Participation (20%): show up, engage, and you’ll be fine• 3 (short) assignments (30%): built to help you become
comfortable with the techniques we discuss in class• 10-minute presentation on a research paper (10%): get
a sense for what’s out there in the Visual Analytics world• Course project (40%)
![Page 36: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/36.jpg)
Course Project• Various industry partners (VIPs) will be pitching
potential datasets and/or analysis questions• You’re also welcome to propose your own!• Goals:- Learn how to break big, unwieldy questions down into
clear, manageable problems- Figure out if/how the techniques we cover in class apply
to your specific problems- Build VA systems to address them
• Several (graded) milestones along the way• Demos and reception on the final day of class
gain real experience | solve real problems build real relationships
![Page 37: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/37.jpg)
What You’ll Get
By the end of this course, you will:• Understand what Visual Analytics is• Know the foundational methods and tools available• Be familiar with some ongoing research in VA• Know how to perform analytical tasks and report your
findings• Have access to guest speakers from really cool places
(IBM, Google, Yelp, and more…)• Have (marketable!) experience developing useful visual
analytics applications for real clients
![Page 38: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/38.jpg)
What I Expect from You
• You enjoy grappling with difficult problems, and you’re excited about “figuring stuff out”
• You are (or are willing to work to become) proficient in programming and debugging• We’ll do crash courses in various tools and environments, but this
is NOT a course in learning general programming techniques• You’re welcome to work in whatever language(s) you prefer, but I’m
more helpful in the ones I know • You’re comfortable asking questions
![Page 39: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/39.jpg)
What You Can Expect from Me• I’m flexible w.r.t. the topics we cover:
- This course is a collaboration- If there’s something you want to learn that’s not on the agenda,
speak up!• I’m happy to share my professional connections:
- If there’s a company or school you’d like to work with for your project, let me know and I’ll reach out to them
- Note: the best way to get a job/internship/etc. is to convince someone on the inside that you’re awesome
• Downside: I have non-standard office hours- Full-time research scientist at MIT; hours at Tufts are limited- My commitment: if you email me during business hours, you’ll get a
response by the end of the day- We can explore other options as needed (Google hangout, etc.)
![Page 40: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/40.jpg)
General Information
• Course website:http://www.cs.tufts.edu/comp/150VIZ
• Syllabus• Textbooks (only one required, free download)• Assignments• Grading• Accommodations
![Page 41: Lecture 01: Introduction](https://reader036.vdocuments.us/reader036/viewer/2022062815/5681690a550346895de01ff1/html5/thumbnails/41.jpg)
Questions?