Modeling Malaysian Public Opinion by Monitoring the Malaysian Blogosphere Brian Ulicny VIStology, Inc. Framingham, MA
Feb 25, 2016
Modeling Malaysian Public Opinion by Monitoring the Malaysian
BlogosphereBrian Ulicny
VIStology, Inc.Framingham, MA
IBlogs GoalsCluster Posts by Event, and Rank using new metrics for • Relevance• Timeliness (& Locality)• Specificity• CredibilityApplied to Malaysian Blogosphere as a Case StudyAFOSR funded; Distributed Intelligence Program
– Automated Ontologically-Based Link Analysis of International Web Logs for the Timely Discovery of Relevant and Credible Information
– Ends November, 2008.
4/2/08 VIStology, Inc. 2
Commercial State of Art
4/2/08 VIStology, Inc.
1. Technorati Blog Search 2. Organized by Date and
Authority.3. Authority = Global Link
Popularity.4. No way to filter blogs
by location currently.5. So must probe via
keywords.6. No aggregation of
topics or links locally.
3
4 VIStology, Inc.
The Malaysian BlogosphereFormer British colony.
English 2nd Most Common Language.Complex ethnic and religious social structure.Press tightly controlled
RSF: 124th of 169.But: Internet growth encouraged.Over 150K Malaysian profiles on Blogger.com.
310K profiles on Facebook.comBlogs have exposed alleged government corruption
and organized protests.Government has intimidated and arrested bloggers.
4/2/08
Malaysian SoPo Blogosphere
4/2/08 5VIStology, Inc.
Malaysia SoPo Blogs:385 blogs in crawl seed yields 220.3K urls, 4693 Sites
@ depth 4• ~2K Blogs
– 1,059 Blogspot Blogs, 801 Wordpress, 16 Typepad• 272 .org, 45 .gov.my • Of the Blogspot blogs (structured profiles)
– 659 Malaysia, 166 other, 234 unspecified• 122 Kuala Lumpur• 88 Selangor (State)• 31 Johor (State), 10 Subang Jaya, 11 Klang, 6 Johor Bahr
– 10 list Koran as favorite book; 4 Bible
4/2/08 VIStology, Inc. 6
Fragmentation of References:Diversity of Links for Bersih Rally Posts
4/2/08 VIStology, Inc. 7
Posting and Linking PatternsMalaysian SoPo Bloggers, Blogging Unfree Press
4/2/08 VIStology, Inc. 8
Cf. 50% of US “A-list”Blog posts contain a news link(Adamic, Glance. 2005)
Clustering by Event:Nov. 10, Kuala Lumpur, Malaysia, police
4/2/08 VIStology, Inc.
StarNST
CNN
BBC
9
Blog Post RelevanceAboutness Determined by Term Frequency
in Post Text plus Linked Text
4/2/08 VIStology, Inc.
IHT article Jeff Ooi PostLinks To
10
Blog Post SpecificitySpecificity Determined by # Unique NPs
in Post Text and Linked Text
4/2/08 VIStology, Inc.
IHT article (16 UNPs) Jeff Ooi Pos(26 UNPs)Links To11
Blog Post Timeliness & Locality:Timeliness is Difference between Post Time and
Dateline
4/2/08 VIStology, Inc. 12
Published: November 11, 2007 Posted by Jeff Ooi on November 11, 2007 10:54 AMTimeliness = 0 days
Dateline: Kuala Lumpur, Malaysia Profile: MalaysiaLocality = 0 miles
Centrality vs. CredibilityBlog EDCGoldTruths
(7.8)Rocky’s Bru(14.8)
ForeverFriends(3.8)
Centrality .546 .472 .012Real* Name Ina Stanley Ahiruddin Attan (Orchid)
Provides Location Durham, NC Kuala Lumpur, MY Banjarmasin, ID
Provides Email No Yes No
Average Posts/Day 0.25 1.9 _
Linking Sites 44 38 1
Avg. Non-Spam Comments
1.2 35 0
Blogging for n Months
5 16 4
4/2/08 VIStology, Inc. 13
Conclusions and Future Work
• 1. It is possible to geolocate many Malaysian bloggers explicitly, particularly Blogspot bloggers.
• 2. System can now extract (nearly all) the features we need automatically now.
• 3. Integration efforts ongoing to tie all the pieces together.
• 4. This will allow us to do a complete, fully automated analysis of important events in the Malaysian (sopo)blogosphere, such as the Bersih Rally example that was repeatedly used in the slides.
4/2/08 VIStology, Inc. 14
16 VIStology, Inc.
IBlogs Architecture
BlogOntology
Event ontology
4/2/08
BaseVISor and Ontology Reasoning
4/2/08 VIStology, Inc. 17
Alias: Pak Lah?
Abdullah BadawiPak LahAbdullah Ahmad Badawi
Bersih Rally in Global Blogosphere:1432 Posts in Week 1
4/2/08 VIStology, Inc. 18
Bersih Rally Posts
4/2/08 VIStology, Inc.
Bersih Rally Hindraf Rally
19