The Strand Map Service Tamara Sumner Computer Science

The Strand Map Service Tamara Sumner Computer Science and ICS This research is funded in part by the National Science Foundation under NSF Awards #226286 and #121724 and under Cooperative Agreements #ATM 9732665 and #ATM-0301213 between the NSF and UCAR, and by NSF Grants SP 035, #0085600, and #0227656.

Educational Digital Libraries: Objectives § New models for the production, distribution, and sharing of learning resources and best practices, with traditional and new partners § Open and interoperable library infrastructure enabling third parties to create new apps and services § Content-rich and adaptive learning environments that build on library collections and infrastructure to positively impact student learning Approach: Contextualization Services Architecture 2

Learning Environments Contextualization Services DLESE Teaching Box Adaptive Knowledge Organization Services Strand Map Service Content-Rich Characterization and Filtering Services Discovery Service

The Strand Map Service § Motivation § Harsh Realities: No Child Left Behind Legislation – 2014, Standards as ‘checklists’ to be covered (Atlas for Science Literacy, AAAS, 2001), Connections between ideas are lost (How People Learn, NRC, 2000) § Opportunities: Standards embody reform agenda, integral to perspective of curriculum reform as a ‘design by assembly’ problem (AAAS, Designs for Science Literacy, 2001) § Partners: UCB, Project 2061 – AAAS, DLESE, UCSB § Goals § ‘Rich representations’ of learning goals and means to visualize relationships between goals § Representations and visualizations available throughout distributed library network 4

AAAS Benchmarks and Strand Maps § Benchmarks: Describe what learners should know, or be able to do, at key stages in their education across the STEM disciplines § Strand maps: Node-link diagrams illustrating how student understanding changes over time 5

What is a Benchmark? 3 -5 When warmer things are put with cooler ones, the warm ones lose heat and the cool ones gain it until they are all the same temperature. Research on the cognitive and scientific basis Research on student (mis-)conceptions Strategies to check student understanding K-2 The sun warms the air, land water. Assessment activities 6

Flow of Matter in Ecosystems Map matter cycle food web plants making food 7

Approach: Knowledge Organization Service components User constituency Visualization interfaces K-12 teachers and students Web service protocol Library developers AAAS information model AAAS staff Related Work: § Concept Maps (Hall et al; O’Donnell et al) § Snap-together Visualizations (North and Schneiderman) § Knowledge Organization Systems (Binding and Tudhope) 8

Research Methodology § Visualization requirements § AAAS Info Space § Visualization Algorithm § Concept Space Interchange Protocol § Benchmarks Repository § Pilot User Study § Deployment Cases 9

Strand Map Service: How it works 10

The Strand Map Service in DLESE 11

CSIP Web Service Protocol § Balance expressiveness with facility § expressiveness -> visual components § facility -> programming walkthrough § Representational State Transfer (REST) § Protocol consists of three services § Service Description § Submit Resource § Query 12

The Strand Map Service in the Harvard. Smithsonian Digital Video Library 13

Reflections and Outcomes § Illustrates educational Cyberinfrastructure § Learning environment based entirely on web services § SMS makes visual components available to developers § Impact on the mindset and practices of our CI partners § Evaluation results – a useful cognitive tool § Controlled study examined influence of interface on cognitive processes of undergraduates § Compared visual interface and keyword-based interface § Students focused on science content, twice as much using visual interface, as opposed to query construction and surface features 14

Evaluation Study Methodology § Compared cognitive strategies using Strand Map Service versus DLESE search interface § 12 participants – undergraduate psychology students § 4 Tasks – teachers using DLESE to find educational resources in order to teach a topic in class § Self-explanation with tasks – coded analyzed verbal protocols “…I’m reading the different little boxes to find out which one sounds more related to why earthquakes occur…this one talks about. . how the core of the earth is hot and that’s what makes earthquakes and volcanic eruptions but. . this one says that they also occur along boundaries with colliding plates, so I’m going to try this one first. . “ “. . I guess the first thing I do is probably the most basic thing, just type in rock cycle into the search engine…And it yielded no results so I’ll go back. . right now I’m just clicking on the options I have here, seeing if I can narrow down the search. . “ 15

Examples of Coded Verbal Protocols 16

Results a b c Mean number of search processes performed during a) formulation, b) performance, and c) evaluation phases 17

Advance Knowledge Sharing (“Concept” only so far!) The Integrative Data Viewer (IDV) and other such tools can access multiple datasets served by GEON and save and share the view using a “Bundle”. An advanced application could allow the user not only to share this visualization bundle of integrated data and model files, but also the concepts that led the user to create a particular bundle. Since each individual dataset is registered to an ontology(ies) at GEON, the user could in principle create a new cross-cutting ontology with associated concepts. An ambitious user could also register the new concept for educational purposes to a strand map. Educator Point of View: (C. Meertens, 20 April 2005) User saves IDV xml bundle, creates associated material that is registered to a strand map GEON Portal IDV user grabs data from several data ontologies (perhaps unrelated as shown) IDV User views all three data sources together and comes to a brilliant integrative science thought! Scientist Point of View: User saves IDV xml bundle, creates a new integrated 18 knowledge concept space, and saves thought in “my GEON”

Further Reading § Butcher, K, S. Bhushan, and T. Sumner (2005). “Multimedia displays for conceptual search processes: Information seeking with strand maps. ” To appear in: ACM Multimedia Systems Journal (Special issue on Multimedia in Education and e. Learning). § Sumner, T. , F. Ahmad, et al. (2005). “Linking Learning Goals and Educational Resources through Interactive Concept Map Visualizations. ” International Journal on Digital Libraries 5(1 (March), Special Issue on Information Visualization Interfaces for Retrieval and Analysis): 18 -24. § Sumner, T. , F. Ahmad, et al. (2004). A Web Service Interface for Creating Concept Browsing Interfaces, D-Lib Magazine, Volume 10, Number 11. § Gu, Q. , F. Ahmad, et al. (2004). Dynamically Generating Conceptual Browsing Interfaces for Digital Libraries using SVG. 3 rd Annual Conference on Scalable Vector Graphics (SVG Open '04), Tokyo, Japan (Sept 7 -10). § Sumner, T. , S. Bhushan, et al. (2003). Designing a Language for Creating Conceptual Browsing Interfaces for Digital Libraries. ACM/IEEE Joint Conference on Digital Libraries (JCDL '03), Houston, Texas (May 27 -31). 19