Quinn by liuhongmeiyes

VIEWS: 1 PAGES: 36

									1. Introduction
2. Rationale & Aim
3. Methodology
4. Results and Discussion
5. Conclusions
6. Limitations and Future Work
Epidemiology
                                    “The study of the distribution and
                                         determinants of health-related
                                         states    or   events   (including
                                         disease)...” [2]




  Hippocrates (ca 460 BC – 370 BC)
  “The father of Western Medicine” [1]
Online Social Networking
• Internet & Online Social Networking (OSN)
• Online social network analysis
• Interaction analysis




    200 Million Users [3]   800 Million Users [4]
•
•
•
Aim
1. Harness and mine user generated content.
2. Explore the behaviour and charactersitics of two distinct
   user groups.

  Investigate User Activity
• Define individual's interaction metrics
• Compare interaction rates of a user / group

  Activity Visualisation
• Disclose interaction patterns
  - who, what, where, when and how users engage
Network Selection & Data Collection
• Facebook.com rational
    Leading OSN in terms of visitor numbers [5]
    An observed phenomenon [6]
    Accessible data and rich user interactions [7,8]

• Data Collection Process
    Stage 1: Profile creation
    Stage 2: User identification
    Stage 3: Data extraction and analysis
facebook.com         Log In                User A



                    1. User B
                    2. User C        Find Classmates /
                    3. User D         friends facility
                     etc, etc



                    Security                 NO
                Access 'Everyone'


                      YES


                    1. User B


                      Wall               Profile Info
                                                         Function Rate
                     Mozilla           Visual Studio
                Firefox generated     Bespoke Parser -   Friend Number
                   source view         User B results


    User A     Export as userb.xml   Data Analysis        Active Period
                       file
 User Analysis
USER ID              00001       00002      00003       00004
Age                     19          22         57          61
Gender                   F          M          M            F
Number of friends      339         732          9          31
Function Total         690         497         18          85
User Comments          46%         42%         0%         21%
Comment Frequency   1.35 days   1.78 days   0.0 days   21.87 days
User replies           21%         31%        65%         24%
Reply Frequency     2.91 days   3.16 days   9.4 days   23.47 days


                   500 User Profiles
   250 Young (15 – 30 )          250 Older ( 50 + )
Measurement Approach:
• Function usage total was derived as a sum for each observed activity
• Normalisation process carried out to a % of each user's activity
• Re-applied to the group as a whole

Activity Frequency
                                            af = n / (x1 – x2)
(activity frequency (af), first date stamp (x1), last date stamp (x2), total number of user activities (n))

•    Calculated as a measure of each individual's total activity count.
•    Used to define the interaction rates of users / groups.
•    Categorise user behaviours.
       • G1 Zero, G2 0-4 days, G3 5-9 days, G4 10-19 days and G5 20 days +
Comment Frequency
                                    Young User      Older Users
 70%


 60%
                 52%   52%
 50%


 40%


 30%
                                         23%
                                                                                      21%
 20%
                                                          15%
                                  10%                                10%   10%
 10%                                                6%
       1%
 0%
            G1               G2                G3               G4               G5
Reply Frequency
                              Young User          Older Users
 70%
                 65%

 60%


 50%


 40%
                                                                          31%
 30%
                       24%                                                           24%
                                       21%                21%
 20%


 10%
                                                  4%                 4%
       2%                         2%
  0%
            G1               G2              G3                 G4              G5
                                                       Activity Count




                                                      10
                                                            15
                                                                    20
                                                                         25
                                                                              30




                                              0
                                                  5
                                    Jan-10


                                    Feb-10


                                    Mar-10


                                    Apr-10


                                    May-10


                                    Jun-10
                                                                                              Visualisation – per year
                                                                                   Comments




                                     Jul-10


                                    Aug-10




Activity Period (per month scale)
                                                                                   Replies




                                    Sep-10


                                    Oct-10


                                    Nov-10


                                    Dec-10


                                    Jan-11
Visualisation – Activity Level
                                                             Comments      Replies
                 10
                           A                             B                                                         C
                  9

                  8

                  7
Activity Count




                  6

                  5

                  4

                  3

                  2

                  1

                  0
                  Jan 10   Feb 10 Mar 10   Apr 10   May 10   Jun 10   Jul 10   Aug 10   Sep 10   Oct 10   Nov 10   Dec 10   Jan 11
                                                        Activity Period (per month scale)
Visualisation – Activity Level
                                                         Activity Count




                                                 1
                                                     2
                                                           3
                                                                  4
                                                                          5
                                                                              6
                                                                                  7




                                             0
                                    01 Aug
                                    08 Aug
                                    15 Aug
                                    22 Aug
                                    29 Aug
                                    05 Sep
                                    12 Sep
                                    19 Sep
                                    26 Sep
                                    03 Oct
                                    10 Oct
                                                                                      Comments




                                    17 Oct
                                    24 Oct
                                                                                                 Weekly Scale (Aug 10– Jan 11)
                                                                                      Replies




                                    31 Oct
                                    07 Nov
Activity Period (per month scale)   14 Nov
                                    21 Nov
                                    28 Nov
                                    05 Dec
                                    12 Dec
                                    19 Dec
                                    26 Dec
1. Individual user behaviours are now able to be isolated and categorised as distinct
   characteristics for each user and each group.
2. Young and older users have distinct characteristics and behaviours that are
   directly associated to the age of the user.
    •   Young users are high activity and high frequency and older users much less
        frequent.
3. As a contemporary approach it is able to now determine behavioural patterns
   which may be applied to identify health risk factors (e.g. social isolation etc),
   applied as a preventive approach.
4. Learning not just what users say but what these patterns of usage disclose about
   a population.
Limitations
• Localised geographical region
• Limited data set


Future Work
• Greater understanding of the behavioural patterns of users
• Extension of user metrics, defining and categorising users types
• Qualitative assessments to determine the relationship between
online and real world user behaviours
References
[1] Strong, W.F.; Cook, John A. (July 2007), "Reviving the Dead Greek Guys", Global Media Journal,
     Indian Edition, ISSN: 1550-7521
[2] World Health Organisation, Health Topics, Epidemiology. Internet:
     http://www.who.int/topics/epidemiology/en/ [06 September 2011]
[3] “Twitter co-founder Jack Dorsey rejoins company”. Internet:
     http://www.bbc.co.uk/news/business-12889048 [06 September 2011]
[4] Facebook. Facebook Statisitcs. Internet: www.facebook.com/press/info.php?statistics [13
     October, 2011].
[5] eBizMBA. Top 15 Most Popular Social Networking sites. Internet:
     www.ebizmba.com/articles/social-networking-websites, October 2011, [13 October, 2011].
[6] K.D. Ala-Mutka et al. “The Impact of Social Computing on the EU Information Society and
     Economy”. JRC Scientific and Technical Report, EUR 24063 EN Available:
     http://ipts.jrc.ec.europa.eu/publications/pub.cfm?id=2819 [12 April, 2011].
[7] A. Mislove, B. Viswanath, K. P. Gummadi, and P. Druschel. "You are who you know: Inferring user
     profiles in online social networks". in Proceedings of WSDM'10, New York, NY, February 2010.
[8] K. Lewis, J. Kaufman, M. Gonzalez, A. Wimmer, and N. Christakis. “Tastes, ties, and time: A new
     social network dataset using Facebook.com,” Social Networks, vol 30, 4, 330-342, October
     2008.
Thank You For Your Time !
       Question & Answers
              Darren Quinn
   Email: quinn-d15@email.ulster.ac.uk
Backup Slides
Data – Age Calculations
User Age Information Scenarios:
 High School and or University Year*
 Date of Birth
 No Information

Calculation Examples:
UID 0012YM, School Year (SY) 2008
**SY – ALA = 1990 (Year of birth)
Current Year (2011) - Year of birth (1990) = 21

UID 0012YM, Uni Year (UY) 2008
**UY – ULA = 1986 (Year of birth)
Current Year (2011) - Year of birth (1986) = 25


Notes:         If both are provided, default to the SY as it is more consistent
               Discrepancies SY and UY do not equal 4 (Either course is longer or shorter)

1 Years are taken as leaving years in all cases
2 Average Leaving Age (ALA)/ University Entry Age (UAE) is 18
3 University Leaving Age (ULA) is 22, Degree course length taken as 4 years
Facebook Statistics 13th October 2011
People on Facebook
More than 800 million active users
More than 50% of our active users log on to Facebook in any given day
Average user has 130 friends

Activity on Facebook
More than 900 million objects that people interact with (pages, groups, events and community pages)
Average user is connected to 80 community pages, groups and events
On average, more than 250 million photos are uploaded per day

Global Reach
More than 70 languages available on the site
More than 75% of users are outside of the United States
Over 300,000 users helped translate the site through the translations application

Platform
On average, people on Facebook install apps more than 20 million times every day
Every month, more than 500 million people use an app on Facebook or experience Facebook Platform on other websites
More than 7 million apps and websites are integrated with Facebook

Mobile
More than 350 million active users currently access Facebook through their mobile devices
More than 475 mobile operators globally work to deploy and promote Facebook mobile products

PALO ALTO, Calif. — Jan. 21, 2011
Facebook today announced it has raised U.S.$1.5 billion at a valuation of approximately $50 billion.
About Facebook
Founded in February 2004, Facebook is a social utility that helps people communicate more efficiently with their friends, family and
coworkers. The company develops technologies that facilitate the sharing of information through the social graph, the digital
mapping of people's real-world social connections. Anyone can sign up for Facebook and interact with the people they know in a
trusted environment.
Product
Facebook, the product, is made up of core site functions and applications. Fundamental features to the experience on Facebook are a
person’s Home page and Profile. The Home page includes News Feed, a personalized feed of his or her friends updates. The Profile
displays information about the individual he or she has chosen to share, including interests, education and work background and
contact information. Facebook also includes core applications – Photos, Events, Videos, Groups, and Pages – that let people connect
and share in rich and engaging ways. Additionally, people can communicate with one another through Chat, personal messages, Wall
posts, Pokes, or Status Updates.
Technology
Facebook is one of the most-trafficked sites in the world and has had to build infrastructure to support this rapid growth. The
company is the largest user in the world of memcached, an open source caching system, and has one of the largest MySQL database
clusters anywhere. The site is largely written in PHP though the engineering team developed a way to programmatically transform
PHP source code into C++ to gain performance benefits. Facebook has built a lightweight but powerful multi-language RPC
framework that seamlessly and easily ties together infrastructure services written in any language, running on any platform. The
company has created a custom-built search engine serving millions of queries a day, completely distributed and entirely in-memory,
with real-time updates. Facebook relies heavily on open source software and releases large pieces of its own software infrastructure
as open source. See http://facebook.com/engineering and http://facebook.com/opensource.
About Facebook

Platform
Facebook Platform is a development platform that enables companies and engineers to deeply integrate with the Facebook website
and gain access to millions of users through the social graph. Facebook is a part of millions of people’s lives all around the world
providing unparalleled distribution potential for applications and the opportunity to build a business that is highly relevant to
people’s lives. More information can be found at <http://developers.facebook.com>.
Privacy, Safety and Security
Facebook has always focused on giving people control over their experience so they can express themselves freely while knowing
that their information is being shared in the way they intend. Facebook's privacy policy is TRUSTe certified, and Facebook provides
simple and powerful tools that allow people to control what information they share and with whom they share it. More information
can be found at http://www.facebook.com/privacy/explanation.php. From its beginning, Facebook has worked to provide a safe and
trusted environment by, for example, requiring that people use their real names. Facebook also works with online safety experts
around the world and has established a global Safety Advisory Board that it consults with on safety issues. More information can be
found at http://www.facebook.com/fbsafety and http://www.facebook.com/security.
Funding
Round one: $500,000 from Peter Thiel, Summer 2004; Round two: $12.7 million from Accel Partners, April 2005; Round three: $27.5
million from Greylock Partners leading the round, Meritech Capital Partners participating, and Accel Partners and Peter Thiel
increasing their investment in the company.
Board
Members: Mark Zuckerberg, Marc Andreessen, Jim Breyer, Don Graham, Reed Hastings, Erskine Bowles and Peter Thiel; Observer:
David Sze; Observer: Paul Madera
Employees
2,000+
Users
Over 750 million active (users who have returned to the site in the last 30 days)
Offices
Headquartered in Palo Alto, Calif. U.S. offices: Atlanta; Birmingham, Michigan; Chicago; Dallas; Detroit; New York; Venice Beach,
Calif.; Washington, DC; Austin International offices: Dublin; Hamburg; Hong Kong; Hyderabad; London; Madrid; Milan; Paris;
Selangor; Singapore; Stockholm; Sydney; Tokyo; Toronto;
Template

								
To top