Big Data: Structured and Unstructured • 259
the diaspora of technical approaches broadly falls into a few groups, which
we have outlined as follows:
1. Statistical (Mathematical) Methods—ese methods all use a num-
ber of mathematical equations and statistical properties of words and
collections of texts. Algorithms that you may nd out more about
on the web that are representative of the mathematical approach
have such names as Latent Semantic Analysis (LSA), Vector Space
Methods (VSM), or Hidden Markov Models (HMM). While these
terms refer to highly technical and detailed recipes for counting and
ranking words and phrases in a text for analysis or indexing, they
essentially do not require any background knowledge to work and so
are fast to ...