Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Interactive decision support system to predict print quality.

Sugani Leman1, Mark R Lehto

  • 1School of Industrial Engineering, Purdue University, West Lafayette, IN 47907, USA.

Ergonomics
|January 30, 2003
PubMed
Summary

This study shows Bayesian inference can diagnose printer defects from customer descriptions. The system accurately identified print issues, improving agent diagnostic capabilities in a simulated call center environment.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Healthcare Professionals Risk Assessments for Alert Overrides in High-Risk IV Infusions Using Simulated Scenarios.

Risk analysis : an official publication of the Society for Risk Analysis·2020
Same author

Classifying injury narratives of large administrative databases for surveillance-A practical approach combining machine learning ensembles and human review.

Accident; analysis and prevention·2016
Same author

Bayesian decision support for coding occupational injury data.

Journal of safety research·2016
Same author

A practical tool for public health surveillance: Semi-automated coding of short injury narratives from large administrative databases using Naïve Bayes algorithms.

Accident; analysis and prevention·2015
Same author

Impact of clinical reminder redesign on physicians' priority decisions.

Applied clinical informatics·2013
Same author

Computerized coding of injury narrative data from the National Health Interview Survey.

Accident; analysis and prevention·2003

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Human-Computer Interaction

Background:

  • Printer users frequently encounter defects like fuzzy images, bands, or streaks.
  • Manufacturers rely on customer descriptions for initial problem diagnosis.
  • Accurate defect identification is crucial for efficient customer support.

Purpose of the Study:

  • To evaluate Bayesian inference for diagnosing 16 types of print defects using customer-provided narratives.
  • To develop and test an interactive decision support system based on a Bayesian model.

Main Methods:

  • Trained a Bayesian model on 1701 narrative descriptions of print defects from 60 subjects.
  • Implemented the model as an interactive decision support system in a simulated call center.

Related Experiment Videos

  • Assessed system performance with 8 agents diagnosing 128 customer-reported print defects.

    • Main Results:

    • The Bayesian model achieved high accuracy, correctly predicting the defect category in 70% of cases (top prediction) and within the top five predictions 94% of the time on the training data.
    • In the simulated call center, the model's top prediction was correct 50% of the time, and within the top five predictions 80% of the time.
    • Agent accuracy in diagnosing print defects improved when utilizing the Bayesian decision support tool.

    Conclusions:

    • Bayesian inference effectively learns from narrative data to accurately classify print defect categories.
    • The developed decision support system shows promise for enhancing customer support in diagnosing technical issues.
    • This approach offers a data-driven method for improving the efficiency and accuracy of technical troubleshooting.