Submitted by Anonymous (not verified) on Thu, 2017-01-19 12:34
Speech with CD or web proceedings
Online reviews of tourism services provide valuable resources of knowledge not only for travelers but also for companies. Tourism operators are more and more aware that user related data should be seen as an important asset. This work-in-progress analyzes free text reviews as well as numerical ratings of group tours with the aim to characterize their relations. This is done with the help of statistical models. On the one hand, these models comprise textual attributes and sentiment scores of the reviews, based on text mining techniques and sentiment analysis respectively. On the other hand, non-textual attributes such as meta data about the tours and user related factors are included. First results imply a very moderate relationship between sentiment scores and ratings; the non-textual attributes appear to have a higher impact.
Information and Communication Technology
ENTER 2015: Volume 6
e-Review of Tourism Research
<u>J. Neidhardt</u>, N. Pobiedina, H. Werthner: <br>"<i><a href="http://3ws1wk1wkqsk36zmd6ocne81.wpengine.netdna-cdn.com/files/2015/02/SP03_ReviewSession_Neihardt.pdf" target="_blank" class="publist">What Can We Learn From Review Data?</a></i>"; <br>Talk: ENTER Conference 2015, Lugano, Switzerland; 02-03-2015 - 02-06-2015; in: "<i>ENTER 2015: Volume 6</i>", e-Review of Tourism Research, (2015), ISSN: 1941-5842; 5 pages.<br><br> <a href="https://publik.tuwien.ac.at/showentry.php?ID=234832&lang=2" class="publist"><i>More information</i></a><br><br>