1 / 21

Date/Time Representation In Support Of Intelligence Analysis

Date/Time Representation In Support Of Intelligence Analysis. David M. Cassel Sarah M. Taylor, PhD Gary J. Katz Lois C. Childs Raymond D. Rimey, PhD Lockheed Martin, IS&S Research Engineering. Outline. How timelines help analysts Requirements for Timelines Lockheed Martin’s research

joy-short
Download Presentation

Date/Time Representation In Support Of Intelligence Analysis

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Date/Time Representation In Support Of Intelligence Analysis David M. Cassel Sarah M. Taylor, PhD Gary J. Katz Lois C. Childs Raymond D. Rimey, PhD Lockheed Martin, IS&S Research Engineering

  2. Outline • How timelines help analysts • Requirements for Timelines • Lockheed Martin’s research • Dachselt & Weiland critiques of standard timelines

  3. How do timelines help analysis? • Timelines provide ordering and temporal context • Make it easier to see conflicts: • “On March 15th (2001 understood from previous context), Mr. Jones left London for Kabul, beginning a trip of several months to countries in South Asia and the Arabian Peninsula.” • “Mr. Smith arrived in London in 2000, establishing a close working relationship with Mr. Jones there during the late spring of the following year.”

  4. Timeline Requirements • Represent various kinds of time expressions (absolute, relative, fuzzy, set) • Reflect ordering and spacing of events • Represent different levels of precision • Represent reliability of information sources • Be able to scale • Maintain user orientation during navigation

  5. Aspect 1: Smooth Zooming • View changes slowly to allow continuity • Hash marks change with zoom level

  6. Aspect 2: Fuzzy Dates and Times • “April ’98” • “until about September 2000”

  7. Aspect 2: Fuzzy Times (continued) <timex val=“2006-06-01T20:15”> 8:15 tonight </timex> <timex2 set=“YES” val=“2006-05-WXX-4> Thursdays in <timex2 val=“2006-05”> May </timex2> </timex2> <timex2 val=“P3SU” anchor_dir=“BEFORE” anchor_val=“2006-06-01”> the past three summers </timex>

  8. Aspect 3: Abstraction • Present different levels of detail • Reflect underlying information

  9. Aspect 4: Context Bar • Shows temporal context • Thumb for slide, resize • Lines within bar

  10. Aspect 5: Stacked Timelines • Shows patterns over time • Visually find approximate patterns • Based on • Week • Month • Year • User defined

  11. Problems with traditional timelines • Dachselt & Weiland identified 6 problems with traditional timelines • Views are discrete and fixed, thus they are bound to one specific [level of detail] and have a static size. • High-level views hide too much data, whereas detailed views suffer from lack of context and orientation. • Scalability, e.g. for mobile devices is not supported.

  12. Problems (continued) • Missing support for zooming lenses, i.e. high-detail views or focus areas within a coarser time view. • Requirement of additional cognitive efforts for reinterpretation and orientation due to missing smooth transition between views. • Display of absolute time is missing, i.e. scroll-bar based views (e.g. in e-mail applications) only show relative position within the collection.

  13. Conclusion • Timelines are helpful for intelligence analysis • But there is room to improve prior tools • We have laid out the features we are developing and shown them to address Dachselt & Weiland’s criteria Thanks for listening! Questions?

  14. References • Dachselt, R.; Weiland, M.:TimeZoom: A Flexible Detail and Context Timeline; Conference on Human Factors in Computing Systems (CHI 2006) Extended Abstracts, April 22-27, 2006, Montréal (Québec, Canada), pp. 682-687 Dachselt & Weiland’s paper • ACE2005: http://www.nist.gov/speech/tests/ace/ace05/doc/ace05eval_official_results_20060110.htm • TERN 2004: http://timex2.mitre.org/tern.html • Ferro, Lisa. 2004. TIDES: 2003 Standard for the Annotation of Temporal Expressions. http://timex2.mitre.org/annotation_guidelines/2003_timex2_standard_v1_3.pdf

  15. Backup

  16. InXight Time Wall • Display: Icons, text • Lanes for multiple timelines • No manipulation of data • Overlapping events are visually blocked • Points in time only http://www.inxight.com/products/sdks/tw/features.php

  17. TildenWoods Centrifuge • Display: text, colored bars • Modes: • Absolute duration • Relative duration • Point in time • Can present multiple timelines • One layer of depth • Must change underlying table http://www.tildenwoods.com/

  18. Visual Analytics VisuaLinks • Display: icons, text • Can compare timelines • Points only • Change database to add/modify material • Icons merged for same type/same time http://www.visualanalytics.com/products/visuaLinks/index.cfm

  19. PNNL ThemeRiver http://www.pnl.gov/infoviz/technologies.html#themeriver

  20. I2 Analyst Notebook • Display: icons, text, links • Not specifically a timeline tool • Does not show time scale, duration http://www.i2.co.uk/Products/Analysts_Notebook/

  21. TimeZoom • Smooth zooming • Good focus feature http://www-mmt.inf.tu-dresden.de/Projekte/Publikationen/details/0604.pdf

More Related