A corpus whose sequence of topics can only be described with memory.
Thirty students each write twelve posts. Each post is about one of three
topics, cooking, astronomy or gardening, and holds seven words drawn with
replacement from the vocabulary of its topic and one everyday word
(night, today, week, friends) shared by all topics. The first two
topics of a student are drawn at random. After two different topics, the
next topic is the third one with probability 0.8 and otherwise drawn at
random; after a repeated topic, it is drawn at random. A first-order model
of the topic sequence cannot represent this rule.
Format
A data frame with 360 rows (one per post) and 5 columns:
- post
Character. The post identifier,
st01_t01tost30_t12.- student
Character. The author,
st01tost30.- turn
Integer. The position of the post among the author's posts, 1 to 12.
- topic
Character. The topic drawn by the simulation.
- text
Character. The text of the post, eight words.
See also
hg_sequences(), which turns a clustering of these posts into
one sequence of topics per student.
Examples
head(forum_posts)
#> post student turn topic
#> 1 st01_t01 st01 1 astronomy
#> 2 st01_t02 st01 2 astronomy
#> 3 st01_t03 st01 3 cooking
#> 4 st01_t04 st01 4 gardening
#> 5 st01_t05 st01 5 cooking
#> 6 st01_t06 st01 6 astronomy
#> text
#> 1 stars nebula galaxy telescope moon nebula comet today
#> 2 stars stars galaxy galaxy stars astronomers astronomers week
#> 3 onions bake garlic oven garlic garlic bake friends
#> 4 watering tomatoes seeds prune soil prune tomatoes today
#> 5 soup salt onions oven soup soup salt today
#> 6 orbit comet comet orbit comet telescope planet week
forum_hg <- text_hypergraph(forum_posts, column = "text", id = "post")
forum_hg
#> Text hypergraph: 360 documents, 34 words (documents as nodes, weight = n)
#> Hyperedges: 34 (words); sizes 48-104, median 63.5
#> doc word count weight
#> st01_t01 comet 1 1
#> st01_t01 galaxy 1 1
#> st01_t01 moon 1 1
#> st01_t01 nebula 2 2
#> st01_t01 stars 1 1
#> st01_t01 telescope 1 1
#> st01_t01 today 1 1
#> st01_t02 astronomers 2 2
#> st01_t02 galaxy 2 2
#> st01_t02 stars 3 3
#> ... 2229 more rows