Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Development of tools for the analysis of messages in controlled social network environments

Garrard, William C. (2017) Development of tools for the analysis of messages in controlled social network environments. Doctoral Dissertation, University of Pittsburgh. (Unpublished)

This is the latest version of this item.

Download (1MB) | Preview


There is sometimes more demand for the attention of healthcare providers than there is supply to go around. This study evaluates a way to make expert mental health social workers more efficient at the task of moderating controlled access social network discussion boards. Sometimes, moderators need to make authoritative posts on these boards known as interventions. These are useful when needed but unnecessary interventions may degrade the benefits of organic discussion. For this study an automated decision aiding system (ADAS) tool was developed which provided the automated analysis and visualization of messages and message sentiment. This tool was designed as a means to make the expert moderators more efficient so more individuals could utilize a discussion board without proportional increase in expert moderators and the associated expense. This study determined that the custom designed automated decision-aiding system had no significant effect on participants determining if messages from such a discussion board are deserving of an intervention response for the measures of accuracy, elapsed time, or judgement confidence. The abstraction of context provided by the ADAS in this study is suspected to explain the lack of significant results, and future work would focus on identifying the level of context supply humans would require for the ADAS to have an effect.


Social Networking:
Share |


Item Type: University of Pittsburgh ETD
Status: Unpublished
CreatorsEmailPitt UsernameORCID
Garrard, William C.wcg5003@gmail.comwcg70000-0002-5141-6239
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairSpring,
Committee MemberKarimi,
Committee MemberHirtle,
Committee MemberRotondi,
Date: 10 May 2017
Date Type: Publication
Defense Date: 27 April 2017
Approval Date: 10 May 2017
Submission Date: 2 May 2017
Access Restriction: No restriction; Release the ETD for access worldwide immediately.
Number of Pages: 160
Institution: University of Pittsburgh
Schools and Programs: School of Information Sciences > Information Science
Degree: PhD - Doctor of Philosophy
Thesis Type: Doctoral Dissertation
Refereed: Yes
Uncontrolled Keywords: Information Science, ADAS, NLP
Date Deposited: 10 May 2017 13:35
Last Modified: 10 May 2017 13:35

Available Versions of this Item

  • Development of tools for the analysis of messages in controlled social network environments. (deposited 10 May 2017 13:35) [Currently Displayed]


Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item