Modeling and Analysis of Intentional And Unintentional Security Vulnerabilities in a Mobile Platform Page: 65
xi, 149 pages : illustrations (chiefly color)View a full description of this dissertation.
Extracted Text
The following text was automatically extracted from the image on this page using optical character recognition software:
7.4.1. Data Collection Procedure
The main requirement for this research is availability of a good data set that includes de-
tails of all the activities in the twitter network such as user profiles, the number of messages in-
terchanged between the users, and the interactions between the users, etc. Since all the online
social networks are based on real individuals, privacy settings make it very difficult to acquire a
proper data set with all activities of each individual from the network. Most popular social network
web-sites provide Application Programming Interfaces (APIs) for running their profile crawlers,
but these are either restricted or need special permissions. However, I circumvented the privacy
restrictions problem by simply programming the Twitter crawler to skip over the restricted users.
Since the number of restricted users is very low compared to the total number of users in Twit-
ter, this strategy is expected to have little effect on the results of the analysis. Another problem I
faced while accessing twitter database is due to Twitter's API Rate Limiting policy which limits
the number of requests per hour for the data records made through the API to 150. Luckily, this
restriction could be waived to white-listed users with special permissions. I obtained these per-
missions from Twitter and could access as many as 20000 records per hour. Table 3 summarizes
the statistics of the twitter database I developed using this process. However, since it is difficult
to visualize such a huge network, I show only the results on 500-node subnet in subsection 7.4.2.
Further the difficulties in hand-labeling the user types forced me to limit the experimentation only
528 randomly sampled tweeter records in subsection 7.4.3.
7.4.2. Results on Link Strength Determination and Context-Dependent Classification
I determined the link strengths of a 500-node twitter network by applying the fuzzy logic
based classification method discussed in section 7.2 on the twitter database developed by us. It
is interesting to visualize some characteristic features of this network from its graphical depiction
in Fig. 7.6. First, it can be observed that the network nodes form strong clusters, and the cluster
structures don't change much when weak links are removed; this indicates that the same set of
tweeters communicate frequently with one another though some are more involved than the other
in tweeting. Next, since it is easy to infer that the tweeters with strong connectivity are close as-
sociates, the classification needs to be applied to the tweeters with low relationship strength. From65
Upcoming Pages
Here’s what’s next.
Search Inside
This dissertation can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Dissertation.
Mohamed Issadeen, Mohamed Fazeen. Modeling and Analysis of Intentional And Unintentional Security Vulnerabilities in a Mobile Platform, dissertation, December 2014; Denton, Texas. (https://digital.library.unt.edu/ark:/67531/metadc700067/m1/79/: accessed July 18, 2024), University of North Texas Libraries, UNT Digital Library, https://digital.library.unt.edu; .