we8there: On-Line Restaurant Reviews

Description

Counts for 2978 bigrams in 6147 restaurant reviews from the site www.we8there.com.

Arguments

Value

we8thereCountsA simple_triplet_matrix of phrase counts indexed by review-rows and bigram-columns.
we8thereRatingsA matrix containing the associated review ratings.

Details

The short user-submitted reviews are accompanied by a five-star rating on four specific aspects of restaurant quality - food, service, value, and atmosphere - as well as the overall experience. The reviews originally appear in Maua and Cozman (2009), and the parsing details behind these specific counts are in Taddy (2011).

References

Maua, D.D. and Cozman, F.G. (2009), Representing and classifying user reviews. In ENIA '09: VIII Enconro Nacional de Inteligencia Artificial, Brazil.

Taddy (2011), Inverse Regression for Analysis of Sentiment in Text. http://arxiv.org/abs/1012.2098

Examples

Run this code

data(we8there)
## use bins to estimate with counts collapsed across equal ratings 1...5
fit <- mnlm(we8thereCounts, we8thereRatings$Overall, bins=5) 
summary(fit)
par(mfrow=c(1,2))
plot(fit, type="reduction", v=as.factor(we8thereRatings$Overall), col=c(2,2,2,3,3))
plot(fit) # type = "response" default

## A small topic model (decrease 'tol' for more precision)
tpx <- topics(we8thereCounts, K=20, tol=10)
par(mfrow=c(1,1))
plot(tpx, group=we8thereRatings$Overall>3, col=c(2,3), labels=c("Bad","Good"))
summary(tpx)