1
|
|
2
|
- Investor Public Relations Navigator SystemTM (patent pend=
ing)
|
3
|
|
4
|
- Getting the message from a news release to a reader or viewer via a =
news
source is a strategic task for public relations practitioners.
- The objective of media relations, quite simply, is to affect media –=
to
get our skillfully crafted message into a news article or story in a
comprehendible way for the reader or viewer.
|
5
|
- We have developed a statistically reliable metric for measuring the
degree to which a particular news article or story impacts a particu=
lar
news release.
- It uses a statistical technique called latent semantic analysis (LSA=
) to
measure semantic relatedness of news releases in corresponding
journalistic articles.
|
6
|
- During the past few decades, cognitive scientists and computational
psycholinguistics developed LSA as a statistical model for comparing
semantic similarity of units of text to each other.
- Originally LSA was designed to improve information retrieval methods=
by
performing retrieval based on derived “semantic” content of words in=
a
query (e.g. “Googling”) as opposed to performing direct word matchin=
g.
|
7
|
- The theoretical assumption of LSA is there is some underlying or
“latent” structure in the pattern of word usage across documents.
- Meanings of a word modify as we use a word in different contexts.
Similarly, different words sometimes have the same meaning dependin=
g on
contexts.
- The idea of latent semantic analysis is that the aggregate of all t=
he
word contexts in which a particular word does, and does not, appear
provides a set of mutual constraints that reflects the similarity of
meaning of words to each other.
|
8
|
- We did four explorations in a study testing the feasibility of the
metric.
- “News Release Flow-Through: News Release/News Article LSA Metric”
Presented at the Association for
Education in Journalism and Mass Communication (AEJMC), Public Relat=
ions
Division, Toronto, August 4, 2004.
|
9
|
- In Exploration 1 we matched news releases and news articles written =
by
journalists about the company. We found that the metric can score
release/story relatedness on underlying semantic structure rather th=
an
word matching.
- The metrics were identical (cosine =3D .76), showing semantic coher=
ence,
but the articles are written differently.
- It initially demonstrates that LSA can provide a reliable metric
measurement of relatedness of the source news release to the target =
news
article.
|
10
|
- Exploration 2 probed the lower-end LSA scoring of semantic relatedne=
ss.
The test was of known unrelated news release.
- With cosines of .45 and .44 the levels are below anticipated minimum
levels of .60.
- Minimally this exploratory test points to the ability of LSA to dete=
ct
non-relatedness of news releases to news articles.
|
11
|
- Exploration 3: LSA can provide a useful relative score, showing the
degree of semantic relatedness of a news release and a news article.=
- From the topic-related news releases/news articles selected, the ra=
nge
(.67 - .85) demonstrated a level of discrimination of the scale.
- This exploratory study had semantically related measures in the .60=
’s,
.70’s and .80’s, tentatively defined as weak, medium and strong
relatedness or coherence respectively.
|
12
|
- Exploration 4 demonstrated a range of LSA cosines from strong impact
(.86), to medium impact (.72) and to weak coherence (.63).
- It is clear from the exploratory study that LSA has the ability to
differentiate levels of coherence, although, because of the limitati=
ons
of an exploratory study, additional research will be needed to verify
and clarify LSA usage, especially the gradation of cosine levels.
|
13
|
|
14
|
|