SentimentAnalyis calculates the sentiment from (tm text corpora based on simple word counts. The package already includes sentiment dictionaries from various sources including e.g. Harvard-IV, Loughran-McDonald and supports user specified ones.
THIS PACKAGE IS UNDER ACTIVE DEVELOPMENT AND NOT READY FOR TESTING - HOWEVER SUGGESTIONS ARE HIGHLY WELCOME !!!
Using the devtools package you can easily install the latest development version of tm.plugin.sentiment from github with
library(devtools)
install_github("mannau/tm.plugin.sentiment")The following lines show the basic functionality calculating the sentiment for the Apple (AAPL) stock:
library(tm.plugin.sentiment)
# retrieve corpus
require(tm.plugin.webmining)
corp = WebCorpus(GoogleFinanceSource("AAPL"))
# score corpus
corp <- score(corp)
sentixts <- metaXTS(corp)
# chart sentiment scores
chartSentiment(sentixts)tm.plugin.sentiment is released under the GNU General Public License Version 3