Information Technology Reference
In-Depth Information
10.3 Generating Metaphors that are Affective
and Effective
The Google n-grams [
4
] is a rich source of popular metaphors of the form
Ta rge t i s
Source
, such as “
politicians are crooks
”, “
Apple is a cult
”, “
racism is a disease
” and
“
Steve Jobs is a god
”. Let
src
denote the set of stereotypes commonly used to
describe a topic
T
, where commonality is defined as the presence of the corresponding
metaphor in the Google n-grams. To find metaphors for proper-named entities, we
also analyze 3/4-grams of the form
stereotype First [Middle] Last
, such as “
tyrant
Adolf Hitler
” and “
boss Bill Gates
”. Thus, e.g.:
(
T
)
src
(
racism
)
={
problem, disease, joke, sin, poison, crime, ideology, weapon
}
src
(
Hitler
)
={
monster, criminal, tyrant, idiot, madman, vegetarian, racist, …
}
Let
typical
denote the set of properties and behaviors harvested for
T
fromWeb
similes (see previous section), and let
srcT ypical
(
T
)
denote the aggregate set of
properties and behaviors ascribable to
T
via the metaphors in
src
(
T
)
(
T
)
. Thus:
(
1
)
srcTypical
(
T
)
=
typical
(
M
)
M
∈
src
(
T
)
We can generate conceits for a topic
T
by considering not just obvious metaphors
for
T
,but
metaphors of metaphors
. Thus:
(
2
)
conceits
(
T
)
=
src
(
T
)
∪
src
(
M
)
M
∈
src
(
T
)
The features evoked by the conceit
TasM
are then given by:
(
3
)
salient
(
T
,
M
)
=[
srcTypical
(
T
)
∪
typical
(
T
)
]∩[
srcTypical
(
M
)
∪
typical
(
M
)
]
The degree to which a conceit
M
is apt for
T
is given by:
)
=
|
(
,
)
∩
(
)
|
salient
T
M
typical
M
(
4
)
aptness
(
T
,
M
)
|
However, we should focus only on apt conceits
M
|
(
typical
M
∈
conceits
(
T
)
where:
(
5
)
apt
(
T
,
M
)
=|
salient
(
T
,
S
)
∩
typical
(
M
)
|
>
0
and rank the set of apt conceits by
aptness
(T, M), as given in (4).
The set
salient
(T,M) identifies the properties and behaviors that are evoked and
projected onto
T
when
T
is viewed through the metaphoric lens of
M
. For affective
conceits, this set can be partitioned on demand to highlight only the unpleasant
aspects of the conceit (e.g. “you are such a baby!”) or only the pleasant aspects (e.g.
“you are my baby!”). The authors of [
22
] further show how n-gram evidence can be
used to selectively project the salient norms of
M
onto
T
.
Search WWH ::
Custom Search