Information Technology Reference
In-Depth Information
Naive Bayes: Apple
Naive Bayes: Microsoft
Maximum Entropy: Apple
Maximum Entropy: Microsoft
70
60
50
40
30
20
10
0
20
21
22
23
24
25
26
27
28
Aug
Sent Date of Tweets
Fig. 2.13
Increase of percentage of positive tweets of Miami on 23 August
Naive Bayes: Twitter
Naive Bayes: Facebook
Maximum Entropy: Twitter
Maximum Entropy: Facebook
70
60
50
40
30
20
10
0
20
21
22
23
24
25
26
27
28
Aug
Sent Date of Tweets
Fig. 2.14
Percentage of positive tweets of London
can be better observed particularly in the cities where a higher number of tweets
are collected such as London and New York. As it can be seen from Fig.
2.14
,even
though the classification method changes, the percentage of positive tweets does not
change much.
As a result, the increase of data; in other words, the increase in the number of
collected tweets, renders the results produced by the system more reliable. In this
way, the system can generate more stable results.
Search WWH ::
Custom Search