Example on how to retrieve data from twitter and do topic model analysis on it.
The analyze_twitter_user R script shows an example on how to fit some topic models on data retrieved from twitter. The data collection is based on the twitteR package. The get_max_tweets script however shows a better method for the data collection and it contains an example on how to download the maximum amount of tweets (~3200) from a user timeline. It is based on the smappR package.
In order to 'do it yourself' you need access to the twitter API and for that you need to create a twitter app (with your twitter account). For that go here . You might also find this tutorial useful.
The fi_stopwords file includes finnish stopwords that are retrieved from here with some manual additions. It is used in cleaning the vocabulary before fitting the topic models.