Christina M. Lirot, Diane Cessna, and Kay H. Connelly - PowerPoint PPT Presentation

1 / 1
About This Presentation
Title:

Christina M. Lirot, Diane Cessna, and Kay H. Connelly

Description:

The first stage of the project will be to develop the software for updating and ... The picture below shows the high-level architecture of our solution, which ... – PowerPoint PPT presentation

Number of Views:46
Avg rating:3.0/5.0
Slides: 2
Provided by: csInd
Category:

less

Transcript and Presenter's Notes

Title: Christina M. Lirot, Diane Cessna, and Kay H. Connelly


1
Christina M. Lirot, Diane Cessna, and Kay H.
Connelly Computer Science Department, Indiana
University Bloomington, Indiana 47405
  • Stage One
  • The first stage of the project will be to develop
    the software for updating and accessing the
    database. The picture below shows the high-level
    architecture of our solution, which consists of
    the following components
  • a PDA program that works with a barcode reader to
    facilitate data entry
  • a web service on the server to which the PDA
    program automatically connects to upload entries
  • a verifier on the server that automatically gets
    entries before updating the database
  • a web service to provide read-only database
    access to 3rd-party applications
  • a user-friendly web site for human access to the
    database

PDA Interface Development
Do we Trust the Information in the Database?
We are currently creating the PDA interface
design in Visual C .Net Framework. We are
developing scanned screen environment and a
manual screen environment for ones who do not
have barcode readers. This will also be
available via the web. There are several screen
shots shown below. This will give you the basics
for manually entering the information.
For many, incorrect nutritional information could
have devastating effects. Diabetics, dialysis
patients and people with allergies are examples
of populations that could both use such a
database, and be harmed by incorrect information.
We have taken steps to ensure that the accuracy
of the data in the database is enough for the
benefits of using our database to outweigh the
risks of receiving inaccurate information.
  • We have identified and evaluated trust models to
    determine which is most appropriate for this
    application.
  • We intend to use trust and reputation scores to
    allow the user to make a more informed decision
    before trusting an entry in the database.
  • We will be combining aspects from trust models
    similar to those used by Ebay and Wikipedia.
  • We will create the user interface in a way that
    minimal text entry is required.

START
Will the Population Make the Entries?
We will investigate the best approach for getting
people to make entries. We will conduct a user
study of a particular motivated population to see
if the motivation of being helped by the product
is enough to elicit participation, or if other
means are necessary. Specifically, we will send a
PDA equipped with a bar code reader and our data
entry application home with diabetics. For the
control group, we will show the participants how
to access the data through the web interface, as
well as provide them with a simple PDA
application that helps them track their nutrition
if the foods they eat are present in the
database. For another group, we will provide an
incentive program as well. Incentives will be
based on the number of entries submitted. For
example, they may be entered into a raffle for a
prize after a certain number of entries. We will
record the number of entries participants make
over a two week period, and monitor their entries
for errors. In this way, we will determine the
type of monetary investment that would be
required to initially populate the database.
The PDA program will allow for very fast entry of
nutritional information and the ability to link
it to the bar code. The web service will allow
multiple entries of the same UPC and will forward
the entries to the verifier. The verifier can
use redundant entries to facilitate automatically
detecting errors and assigning trust levels to
data. For example, if an item has been entered 5
times with the same values, the data would be
deemed highly trustworthy whereas if an item has
been entered twice and has conflicting data, the
data would have a low trust value.
References
  • Kay Connelly, Katie A. Siek, Yvonne Rogers,
    Josette Jones, Michael A. Kraus, Susan Perkins,
    Laurie L. Trevino and Janet L. Welch, Designing a
    PDA Interface for Dialysis Patients to monitor
    diet in their everyday Life. In the Proceedings
    of HCI International 2005.
  • English, Colin, Nixon Paddy, et al. Dynamic Trust
    Models for Ubiquitous Computing Environments.
    Department of Computer and Information Sciences,
    University of Strathclyde presentation 2004.
  • S.S. Intille, C. Kukla, R. Farzanfar, and W.
    Bakr, Just-in-time technology to encourage
    incremental, dietary behavior change, in
    Proceedings of the AMIA 2003 Symposium Wiley
    2003.

Stage Two
The second stage of the project is to seed the
database with enough entries to demonstrate its
usefulness to the public. In addition to
manually entering data for a specific category of
food (i.e. breakfast cereals), we will
investigate how to encourage specific populations
to participate in seeding the database by
providing incentives. A basic incentive is the
ability to use the database through the web-site
and/or through a nutrition monitoring
application. A more obvious incentive is a chance
to win a 100 gift certificate for every 30 valid
entries.
http//mypage.iu.edu/dcessna/CREU
Write a Comment
User Comments (0)
About PowerShow.com