Experiments run on the first set of terms (generic)
resulted into three groups, according to the size of
word cloud generated, with respect to the values of
the number of topics, number of top words and the
probability threshold. These groups are: a) Group A
– a small scale group, b) Group B – a medium scale
group and, c) Group C – a large scale group.
Group A produced a total of 44 words, group B
554 words, and group C 921 words. Comparing the
top words of group A against the top words of the
other two groups (Figure 3) it can be argued that the
groups are well separated.
Group C produced more content than the group
B, which produced more content than group A.
Moreover, group C’s produced content is
characterized by more variety in contrast to the other
two groups and the top words of the content
produced, in terms of occurrences, in the small-scale
case are top w