idm (version 1.3.1)

tweet: twitter data set

Description

The data set refers to a small corpus of messages or tweets mentioning seven major hotel brands. It was gathered by continuously querying and archiving the Twitter Streaming API service, using the twitteR package in R. A total of 7,296 tweets were extracted within a time period of 6 days, from June 23th to June 28th 2013. Only tweets in the English language were considered. A sentiment polarity variable was calculated, indicating the sentiment value of each message and a third variable, user visibility or popularity, as measured by the number of followers each user had, was also included in the dataset

Usage

data("tweet")

Format

A data frame with the following variables:

Brand

The hotel brand mentioned in the tweet: 1=Hilton, 2=Intercontinental, 3=Marriott, 4=Bestwestern, 5=Starwood, 6=Hyatt, 7=Choice

Sentiment

Sentiment for each tweet: 1=negative (-), 2=mixed (+/-), 3=positive (+), 4=very positive (++)

UserVis

User popularity/visibility in Twitter: 1=low, 2=medium, 3=high

References

Iodice D' Enza, A., & Markos, A. (2015). Low-dimensional tracking of association structures in categorical data, Statistics and Computing, 25(5), 1009-1022.

Examples

data(tweet)