|
|
 |
Content Analysis
Examples
The examples shown below,
were obtained with the basic NLP
functions of
our engine,
on unstructured text, and
without the use
of a semantic
network.
HTML tags
were ignored.
450 files of
the skepdic.com
dictionary were processed.
To increase the
difficulty of the content analysis,
some of the following words were used
as follows:
will is
used as a proper noun, noun,
and modal.
can is
used as a noun
and modal.
may is used
as a proper noun, noun, and modal.
are is
used as a noun and verb.
at is used as a noun
and preposition.
to is used
as a preposition and an adverb.
as is
used as a noun, preposition, conjunction, and adverb.
an is
used as noun and determinant.
The use of the semantic
network, which is
currently under construction,
will improve the accuracy of
the analysis.
Our next test run
( with
or used
as a verb, adjective, noun, preposition, and conjunction;
and
used as a verb, noun, and conjunction;
in used
as a verb, adjective, noun, and preposition ),
will
be executed, once the semantic
network
is completed,
on the
~650,000 files
of the
Wiki encyclopedia.
The following words are only tracked,
when they are
the actual subject of the page.
| term, name, word,
expression, subject, |
| kind, type, form,
example, |
| object, thing, other,
others, |
| concept, idea,
notion, theory, |
|
