Figure - uploaded by Matthew J. Schneider
Content may be subject to copyright.
Download
View publication
Summary statistics for predictors of review usefulness.

Summary statistics for predictors of review usefulness.

Source publication
Figure 1: Proposed authorship attribution method.
Figure 2: Second stage accuracy results for the Yelp dataset.
Figure 3: The number of candidate authors by amount of training data...
Predictor variables calculated from the text of each document.
+4 Summary statistics for predictors of review usefulness.
The Privacy of Online Reviews
Preprint
Full-text available
  • Aug 2020
Online reviews are an important source of information on products and services for consumers and firms. Although incentivizing high-quality reviews is an important business objective for any review platform, we show that it is also possible to identify anonymous reviewers by exploiting the characteristics of posted reviews. Using data from major re...
Cite
Download full-text

Contexts in source publication

Context 1
... Ipeirotis (2010), we use regression to explain review usefulness with a variety of standard features ( Singh et al. 2017), such as the Flesch-Kinkaid reading level, a measure of the readability of the review. Summary statistics for the variables used appear in Table 6. We previously discussed all variables with the exception of polarity (e.g., sentiment or tone), which is measured using a weighted average of words that are tagged by their sentiment (positive, negative, or neutral) according to the dictionary in Hu and Liu (2004). ...
Context 2
... and 2nd Stage results are based on Table 3, Figure 3, and a Sophisticated Platform with 100 training documents. Table 7. We previously discussed all variables with the exception of polarity (e.g., sentiment or tone), which is measured using a weighted average of words that are tagged by their sentiment (positive, negative, or neutral) according to the dictionary in Hu and Liu (2004). ...