Automaatika referaat (eng)
and then selects the one that explains the most variance.
In the second approach, the evaluation of content may be accomplished through
the specification of vocabulary (i.e., the evaluation of the other aspects of writing is
performed as described above). Latent Semantic Analysis and its variants are employed
by some developers to provide estimates as to how close the vocabulary in the candidate
answer is to a targeted vocabularly set.
The third approach is to develop models that are based on a "gold standard"
formulated by experts. To date this mechanism for developing models is more theoretical
than applied. If normative models can be created for the relevant dimensions of writing
(i.e., age and writing genre), then other variables could be tailored on an a priori basis to
generate a statistical blueprint for the ideal response. The blueprint may or may not be
aligned with human ratings
annaabi.ee 
