Information Technology Reference
In-Depth Information
Fig. 5.3
Frequency of
tweeting as to time and
persona. The columns of this
table represents the persona
and the row of the table
represents the time from
Monday to Sunday. The
content of each cell represents
the times of tweets in 1 hour
Fig. 5.4
The relationship of
the similarity and the number
of pairs. This result shows
that there exists small number
of pairs of very high
similarity between users
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0
5000
10000
15000
20000
25000
30000
Ranking of the pair
This problem is not restricted to the similarity of lifestyle. Activity BOTs and the
technologies that imitate human conversation have become sophisticated. After
examining such kinds of BOT technologies, we concluded that clever BOTs would
skew the data in any statistical approach. Such kinds of problems would still be hard
to avoid even when using authentication technologies because some actual person
or company authenticates most of the BOTs.
5.6 Trust by Meta-Communication
The most serious shortcoming of the above approach is that we considered trust to
be represented by statistical information over the social relationship. The human
race, over time, has designed various social methods and social devices to establish
trust. We have to reconfirm such kinds of deep knowledge about trust. A trust rating
method requires more precise evidence of trust.
