 |
 
|
|
|||||||||||||
|
Introduction EarthTutor is an Intelligent Tutoring System (ITS) that integrates with the NIH's ImageJ and NASA's Image2000 image processing applications to provide an interactive and adaptive remote sensing learning experience. EarthTutor provides software-based labs that instruct in exciting new ways by leveraging emerging artificial intelligence techniques. EarthTutor presents earth science problems to students (e.g., "What vegetation grows near your school?", "How has the ozone hole changed in size over the last 20 years?", "Where does new ice form in the Antarctic?") then guides students to solve these problems with the imagery and image processing tools at their disposal, allowing students to direct the interaction.
We are actively seeking additional pilot volunteers. EarthTutor's pilot program is currently underway. If you are an educator or student and would like to participate in the program, please contact Aaron Bell. If you are already involved in the pilot and would like to download the EarthTutor beta or obtain more information please visit the pilot page. Frequently Asked Questions
Q: What is an Intelligent Tutoring System (ITS)? An ITS provides customized instruction to students as they interact with a software program or simulation. The goal of ITS software is to offer the benefits of a one-on-one human instructor. EarthTutor interacts with students as they work with image processing software and earth science datasets. Labs developed for EarthTutor combine lesson content (such as HTML web pages) with activities and questions. In each lab students learns to measure, calibrate, color, slice, plot and otherwise process and analyze earth science imagery. During the activities, EarthTutor monitors students closely as they work, which allows it to provide immediate feedback that is customized to a particular student's needs. As the student moves through the labs, EarthTutor assesses the student and tailors the presentation of the content to a student's demonstrated skill level. Please see a walk-through example of the EarthTutor ITS in action. Q: With which image processing programs does EarthTutor integrate? EarthTutor has been developed as a plug-in for the NIH's ImageJ and NASA Image2000 (NI2k). Both of these free image processing software packages are intended for both the classroom and scientific research settings. NASA Image2000 is specialized satellite imagery analysis software, while ImageJ is more generalized and can be applied to many domains (medicine, chemistry, etc.) For further info on ImageJ, including download, visit the ImageJ homepage. ImageJ is compatible with Mac, Windows and Linux. For further info on NASA Image2000, including download, visit the NASA Image2000 homepage. NASA Image2000 is compatible with Windows. Q: What benefits does EarthTutor offer relative to existing paper or web-based remote sensing lab materials? Earth science classes in colleges and high schools use a variety of satellite image processing software to teach earth science and remote sensing principles. However, current tutorials for image processing software are often paper-based or lecture-based and do not take advantage of the full potential of the computer context to teach, immerse, and stimulate students.
Q: Who is the audience for EarthTutor? The lab materials being produced for EarthTutor are intended for high school and undergraduate earth science students. The goal of EarthTutor is to provide a simple way for educators to include satellite imagery in their labs while accommodating students at varying skill levels and from diverse backgrounds. Q: What materials will be packaged with EarthTutor? EarthTutor will be packaged with a library of courses, including:
We are interested in producing a wide range of remote sensing content for EarthTutor and explore the exciting educational opportunities made possible by this novel medium. If you or your organization is interested in developing content for EarthTutor, please contact us. EarthTutor will also include an authoring tool which will allow educators to quickly build localized EarthTutor content for their nearby geographic region or other regions of interest. Q: How are new labs for EarthTutor produced? The EarthTutor system includes an authoring tool that allows course developers to create new courses or modify existing courses. The tool is basic enough to allow teachers to customize existing labs, but powerful enough to allow professional producers to create highly-interactive labs that engage students in unprecedented ways. Q: Can you provide a sample walk-through demo of EarthTutor? A student starts by launching ImageJ. ImageJ is an image processing application that supports standard image formats and enables to students process and analyze imagery using typical image processing tools. To begin a lab, the student chooses "Run EarthTutor Lab..." from ImageJ's Plugins menu.
EarthTutor displays a welcome window that prompts the student for his login name or group name and password. (EarthTutor can be configured to read and write student data to and from a central shared machine, or it can be setup to run entirely on a single machine.) Since EarthTutor's courses are generally completed in multiple sessions, students must have an account so that their history can be loaded.
Next, EarthTutor presents a list of available courses with descriptions, along with the student's progress in each course. EarthTutor is packaged with 4 courses: oceanography, global vegetation, natural disasters and polar ice. (The pilot version contains only Oceanography, but additional courses are in production.) Students can complete courses in multiple sittings. The authoring tool allows educators to build new courses.
When a student selects a course, the student is then prompted to select a lab within the course. Each course typically consists of 3-7 labs. Labs are designed to require about 45 minutes for the student to complete. For this walkthrough, assume the student opened the oceanography course. The oceanography course is comprised of six labs.
After the student selects a lab, the EarthTutor instruction panel opens, which contains multimedia course materials and interactive activities and questions. EarthTutor labs are designed as a stack of cards which students progress through sequentially. The instruction panel shows the contents of one card at a time. The instruction panel updates in concert with the students interactions with ImageJ. The first card of the oceanography course is shown below, which presents the context and significance of the topic the lab is about to cover. Generally labs begin with an overview card of this sort.
Early labs in each course familiarize a student with the basic dataset and image processing tools. The first oceanography lab introduces sea surface temperature (SST) data as recorded by the AVHRR sensor. This lab teaches students how to color, calibrate and scale SST images, which enables students to better understand what they are seeing. Because this may be the first time a student is exposed to these concepts, EarthTutor provides detailed instruction on how to perform each image processing task.
As students work through activities and answers question, EarthTutor provides immediate feedback. If a student makes a mistake, EarthTutor can provide a hint to help the student along. Hints may be provided as text balloons or helpful overlays on the working image. For example, in the first oceanography lab, EarthTutor provides a hint if a student has trouble locating the warmest temperatures on the image.
EarthTutor presents many questions to the students to ensure they comprehend the materials and to gauge the areas in which they may need additional help. The tutor can ask "image interactive", multiple choice, fill-in-the-blank, tabular, and essay questions. Hints may be associated with all questions and a student's particular responses. EarthTutor provides immediate feedback when a student answers a question, with the exception of essays, which are saved for a teacher to evaluate after a lab is complete. A screenshot is provided from the first oceanography lab that asks the students a multiple choice and essay question about "temporal resolution".
Students must generate spatial and temporal plots from imagery and make interpretations. In the oceanography module, students must create and compare vertical and horizontal profiles of sea surface temperatures.
As a student advances, he or she will be expected to apply the image processing skills learned he or she has learned. EarthTutor's coaching begins to "fade", or become less prevalent. Students must begin anticipating when to apply certain processing or analysis techniques. They must also remember the steps followed to correctly apply them. At these later stages, EarthTutor by default will not present rote instruction, unless the student explicitly asks for instruction or the student makes a mistake. Forced to recall what they have learned, students better assimilate the general concepts.
In later labs in a course, after the student has become comfortable with ImageJ, content delivery begins to revolve more around earth science concepts than image processing skills. In the oceanography course, later labs cover upwelling, relating wind data to sea surface temperature data to biological response.
At the end of each lab, the student is presented with a report card summarizing his or her answers. Teachers can also view any student's report card using the Learner Management System (LMS).
At the end of some labs, EarthTutor suggests a topic for follow-up study and points a student to relevant web pages and datasets. The student is encouraged to independently explore the skills he or her just learned. For example, the oceanography lab in which students observe temporal variation in SST over the seasons of the years refers students to yearly data in which they can observe El Niņo events.
EarthTutor's pilot program is currently underway. For more information on this, please refer to the top of this page. A publicly-available release of EarthTutor will be available in December 2005 which will include the four earth science courses outlined above. The channels of distribution are to be determined. Please stay tuned. Q: Who has contributed to EarthTutor? EarthTutor's labs are being developed by a team of remote sensing experts including Prof. Kevin Arrigo of Stanford University, Prof. Chris Van de Ven of Albion College, and Prof. Conghe Song of the University of North Carolina. Many EarthTutor labs are being adapted from NASA Goddard's Studying Earth Environment From Space educational package, with great help from Elizabeth Smith, Asst. Research Professor, of Old Dominion University. Q: Which AI (artificial intelligence) technologies does EarthTutor employ? EarthTutor's adaptive approach is based on emerging AI research. Bayesian networks are employed to model a student's proficiency with different earth science and image processing concepts. Agent behaviors are used to track the student's progress through activities and provide guidance when a student encounters difficulty. Q: Who do I contact for more information? Questions and comments regarding EarthTutor should be directed to Aaron Bell (EarthTutor Project Manager) by email at bell@stottlerhenke.com. Questions regarding Intelligent Tutoring Systems and related research can be directed to Dr. Sowmya Ramachandran at sowmya@stottlerhenke.com. Many thanks to Kristen Parton (Lead Developer) and Art Reis (Content Engineer), who have made EarthTutor come to life! More information about Stottler Henke and our other ITS projects can be found on our website at www.stottlerhenke.com. |
The aim of this NASA Earth Science Enterprise-funded project is to foster the use of satellite data in science classrooms. EarthTutor has Phase II funding under the NASA Small Business Innovative Research program (contract #NNA04CA11C).
 |
|
 |
|
Copyright © 2005 Stottler Henke, Inc. All rights reserved. | |