Let us consider as an example a Spanish sentence and the meaning
representation (TMR) produced by
K for that sentence. The following sentence is taken from a news article on the acquisition of a pharmaceutical company by the Roche group:
(1) El grupo Roche, a través de su compañía en España, adquirió Doctor Andreu, se informó hoy aquí.
When translated into English, the sentence reads:
The Roche group, through its company in Spain, acquired Doctor Andreu, it was announced here today.
Figure:
Partial text meaning representation (TMR) of sentence (1).
Given this input, the TMR produced as output by the
K system is shown in figure
.
The meaning of sentence (1) is comprised of an INFORM-132
event carried out by an unknown HUMAN-131 agent. The theme
of the INFORM-132 event is an ACQUIRE-129 event where the agent,
ORGANIZATION-126, named ``grupo Roche'' acquired the theme,
ORGANIZATION-130, named ``Doctor Andreu.'' This acquisition was done
through the instrument, CORPORATION-127, which is located in the
NATION-128 named ``España.'' In other words, the meaning of the
sentence as represented in the TMR is:
Someone informed that the organization named ``grupo Roche'' acquired the organization named ``Doctor Andreu'' through the corporation located in the nation named Spain.
It may be noted that symbols such as INFORM, ACQUIRE, AGENT, THEME,
and so on are, in fact, the concepts that are classified in the
ontology. The TMR is produced by instantiating these concepts in the
ontology and linking them together according to the relationships
between the meanings contained in the input text. We will return to
this example later in this article to illustrate how the meaning
representation was produced by the
K system.
Kavi Mahesh