A Machine Learning Approach for Rating the Quality of Depression Treatment Web Pages

Zhang, Yanjun; Cui, Hong; Burkell, Jacquelyn; Mercer, Robert E.

Permalink

https://hdl.handle.net/2142/47314 Copy

Description

Title

A Machine Learning Approach for Rating the Quality of Depression Treatment Web Pages

Author(s)

• Zhang, Yanjun
• Cui, Hong
• Burkell, Jacquelyn
• Mercer, Robert E.

Issue Date

2014-03-01

Keyword(s)

• machine learning
• Naive Bayes classification
• quality assessment
• web health care information
• evidence-based health care guidelines

Abstract

As health care information proliferates on the web, the content quality is varied and difficult to assess, partially due to the large volume and the dynamicity. This paper reports an automated approach in which the quality of depression treatment web pages is assessed according to evidence-based depression treatment guidelines. A supervised machine learning technique, specifically Naive Bayes classification, is used to identify the sentences that are consistent with the guidelines. The quality score of a depression treatment web page is the number of unique evidence-based guidelines covered in this page. Significant Pearson correlation (p<.001) was found between the quality rating results by the machine learning approach and the results by human raters on 31 depression treatment web pages in this case study. The semantic-based, machine learning quality rating method is promising and it may lead to an efficient and effective quality assessment mechanism for health care information on the Web.

Publisher

iSchools

Series/Report Name or Number

iConference 2014 Proceedings

Type of Resource

text

Language

english

Permalink

http://hdl.handle.net/2142/47314

DOI

https://doi.org/10.9776/14065

Copyright and License Information

Copyright 2014 is held by the authors of individual items in the proceedings. Copyright permissions, when appropriate, must be obtained directly from the authors.

Owning Collections