Pro-Active Detection of Content Quality in TurboTax AnswerXchange
I. Podgorny, M. Cannon, and T. Goodyear. Proceedings of the 18th ACM Conference Companion on Computer Supported Cooperative Work and Social Computing (CSCW '15 Companion), Vancouver, Canada, page 143--146. New York, ACM, (2015)
DOI: 10.1145/2685553.2698992
Abstract
User satisfaction in social question-and-answer (Q&A) systems depends on the quality of answers typically measured by a proxy metrics of user votes on the answers. We show that user votes in TurboTax AnswerXchange (AXC) can be predicted with reasonable accuracy based on the attributes of the question alone. This provides an opportunity for 'pro-active' detection of potentially high or low quality content in real time while the question is still being formulated. As a result, undesirable content can be prevented by instructing the user to re-phrase the question. We can also optimize the AXC answer queue or tweak the AXC point system to generate higher quality answers.
%0 Conference Paper
%1 PodgornyCannonGoodyear15CSCW
%A Podgorny, Igor A.
%A Cannon, Matthew
%A Goodyear, Todd
%B Proceedings of the 18th ACM Conference Companion on Computer Supported Cooperative Work and Social Computing (CSCW '15 Companion), Vancouver, Canada
%C New York
%D 2015
%I ACM
%K 01624 paper ai social software answer
%P 143--146
%R 10.1145/2685553.2698992
%T Pro-Active Detection of Content Quality in TurboTax AnswerXchange
%X User satisfaction in social question-and-answer (Q&A) systems depends on the quality of answers typically measured by a proxy metrics of user votes on the answers. We show that user votes in TurboTax AnswerXchange (AXC) can be predicted with reasonable accuracy based on the attributes of the question alone. This provides an opportunity for 'pro-active' detection of potentially high or low quality content in real time while the question is still being formulated. As a result, undesirable content can be prevented by instructing the user to re-phrase the question. We can also optimize the AXC answer queue or tweak the AXC point system to generate higher quality answers.
%@ 978-1-4503-2946-0
@inproceedings{PodgornyCannonGoodyear15CSCW,
abstract = {User satisfaction in social question-and-answer (Q\&A) systems depends on the quality of answers typically measured by a proxy metrics of user votes on the answers. We show that user votes in TurboTax AnswerXchange (AXC) can be predicted with reasonable accuracy based on the attributes of the question alone. This provides an opportunity for 'pro-active' detection of potentially high or low quality content in real time while the question is still being formulated. As a result, undesirable content can be prevented by instructing the user to re-phrase the question. We can also optimize the AXC answer queue or tweak the AXC point system to generate higher quality answers.},
added-at = {2017-05-24T11:19:45.000+0200},
address = {New York},
author = {Podgorny, Igor A. and Cannon, Matthew and Goodyear, Todd},
biburl = {https://www.bibsonomy.org/bibtex/2fc5aba1b59b3f7cba47a364c97436a6d/flint63},
booktitle = {Proceedings of the 18th ACM Conference Companion on Computer Supported Cooperative Work and Social Computing (CSCW '15 Companion), Vancouver, Canada},
doi = {10.1145/2685553.2698992},
file = {ACM Digital Library:2015/PodgornyCannonGoodyear15CSCW.pdf:PDF},
groups = {public},
interhash = {34f733956b8a49b82817ecf18ba6dbe4},
intrahash = {fc5aba1b59b3f7cba47a364c97436a6d},
isbn = {978-1-4503-2946-0},
keywords = {01624 paper ai social software answer},
pages = {143--146},
publisher = {ACM},
timestamp = {2017-07-13T18:13:31.000+0200},
title = {Pro-Active Detection of Content Quality in {TurboTax AnswerXchange}},
username = {flint63},
year = 2015
}