Multidimensional Patterns of Disturbance in Digital Social Networks

RWTH Aachen University Multidimensional Patterns of Disturbance in Digital Social Networks Dimitar Denev Lehrstuhl für Informatik V Information Systems Prof. Dr. Matthias Jarke Lehr- und Forschungsgebiet Knowledge-based Systems Prof. Gerhard Lakemeyer Ph.D. Advisors: Ralf Klamma Marc Spaniol Master Thesis Final Presentation

Agenda Motivation Problem Analysis Approach State of the Art Model of Digital Social Networks Pattern Language PALADIN Conclusions and Outlook

Motivation Trolls – persons who post only in threads, started by themselves Context Yahoo! Mailing list „Greek Mythology Link“ Discussion about the movie „Troy“ Message of a troll Troy is a MOVIE – message containing deliberate error Movies are current mythology – message posted as a generally accepted fact without a proof or analysis Is Christianity and all that other stuff myth, history, religion or what – inflammatory message including a contemptuous comment on religious thematic.

Problem Statement Disturbance as a new source of information and a starting point for learning processes Hinders the communication in the network Compels individuals to leave the network Difficulties for the disturbances to be discovered or predicted Multidimensional context of the digital social networks Large size of the networks Knowledge about the disturbances is mostly from experience and observation

A pattern language overcomes the difficulties for discovering and describing disturbances Pattern – a general repeatable solution to a commonly recurring problem [Alexander 1978] Machine-readable description of the patterns - XML-based Pattern Language for Multidimensional Disturbances Automatic Analysis of digital social networks for disturbances with the pattern language Solution Approach

Solution Approach The model of the digital social networks is a based on Actor-Network Theory (ANT) Graph Representation Social Network Analysis (SNA) I* Framework Multidimensionality of the digital social networks reflected in the model Sociology Computer Science Media Theory Graph Theory Social Capital Theory

State of the Art Digital Social Networks Projects Relations built on the information from Google, Friend-Of-A-Friend network, Bibliography Dependencies derived from the technical dependencies Posting in the same thread Relations Social Network Analysis, Semantic Web Individuals Friend-Of-A-Friend network, Google results Flink [Mika 2005] Temporal Analysis Developers, Software Components Eclipse IDE, CVS Repository Ariadne [de Souza et al. 2004] Social Network Analysis, Statistics Individuals, Mails, Threads, Genres Mailing List COMB [Boudourides et al. 2002] Analysis Approach Actors Media

Actor - the basic unit of the model, no difference between technical and social actors. Semantics, given to the actors from the interpretation in the context of digital social networks: Member – any person or group, part of the digital social network Medium – an actor which enables the members to exchange information Artefact – objects created by the members using some medium Relation – a relation between two actors Network – set of actors along with their relations Model of Digital Social Networks Actor-Network Theory [Latour 1997]

Digital Social Network Model of Digital Social Networks Digital Media I* Dependencies Members Artefacts Member Network

Member types defined according to patterns of behavior Answering Person Questioner Troll Spammer Conversationalist Member properties , defined with the help of SNA Centrality types: degree centrality, closeness centrality, betweenness centrality - determined by the position of the member in the network Efficiency – describes the existence of structural holes Model of Digital Social Networks Members

Medium – an actor which enables the members to exchange information Every network supports a set of media A medium affords the creation of a certain set of artefacts Media types Email Discussion group Chat room Blog Wiki Transaction-based web sites URL Model of Digital Social Networks Media

Artefact – objects created by the members using some medium Artefact types Message Burst Thread Blog entry Comment Conversation Feedback (Rating) Artefact properties – author, date of creation, reply to Model of Digital Social Networks Artefacts

I* Dependency types Goal Resource Task Soft goal Dependencies in digital social networks Structural dependencies Communication dependency Cross-media dependencies Coordination dependency Artefact dependency Model of Digital Social Networks I* Framework [Yu et al. 1997]

Network Coordinator Gatekeeper Hub Member Iterant Broker URL isA isA isA Coordination Artefact Communication Model of Digital Social Networks I* Dependencies Example isA

State of the Art Pattern Languages Projects „ Asynchronous collaborative learning“, „Student group management“ no patterns available „ Working in small groups“, „Overlapping responsibilities“ „ Citizen access to simulations“, „Online Community Service Engine“ Pattern Examples XML Schema Synopsis, Problem, Context, Forces, Rationale, Pattern Link Human-Computer Interface PLML [Fincher 2004] Not available Not available Not available Formal Definition Problem, Analysis, Solution, Context e-Learning E-LEN [Steeples et al. 2004] Essence, Context, Discussion, Implication, Pattern Relations Computer-Supported Collaborative Work PoInter [Viller et al. 2000] Problem, Context, Discussion, Solution Social Studies Public Sphere Project [Schuler 2002] Pattern Structure Domain

Pattern – a general repeatable solution to a commonly recurring problem [Alexander 1978] Pattern structure Disturbance Forces and force relations Solution Rationale Dependencies Pattern relations Pattern Language Pattern Structure

Variables – simple variables ( troll, thread ), properties ( thread.author ) and set variables ( v 1 ,…,v n ). Operations Arithmetic (+, -, *, / ) Aggregate ( SUM , COUNT , AVERAGE ) Logical (&, |, ~, FORALL and EXISTS ) Comparison ( = , != , > , < ). Rules for variable binding Simple variables – pattern parameters, actors or set variables Properties – actor properties or relations Set variables – actors Interpreted by a finite state automaton Pattern Language FELP

Troll Pattern : This pattern tries to discover the cases when a troll exists in a digital social network. A troll in the network is considered a disturbance. Disturbance : (EXISTS [medium | medium.affordance = threadArtefact]) & (EXISTS [troll |(EXISTS [thread | (thread.author = troll) & (COUNT [message | (message.author = troll) & (message.posted = thread)]) > minPosts]) & (~EXISTS[ thread 1 , message 1 | (thread 1 .author 1 != troll) & (message 1 .author = troll & message 1 .posted = thread 1 ]))])]) Forces : medium; troll; network; member; thread; message; url Force Relations : neighbour(troll, member); own thread(troll, thread) Solution : No attention must be paid to the discussions started by the troll . Rationale : The troll needs attention to continue its activities. If no attention is paid, he/she will stop participating in the discussions. Pattern Relations : Associates Spammer pattern. Pattern Language Sample Pattern

v 1 ,...,v n – variables bound to actors a 1 ,...,a n p 1 ,…, p m – pattern parameters d – disturbance with d=(v 1 ,...,v n , p 1 ,…, p m ). μ 1 ,…, μ m – substitutions for the pattern parameters Set Pattern Parameters : d = d(v 1 ,...,v n , p 1 / μ 1 ,…, p m / μ m ) Pattern Language Algorithm for Pattern Application 1. Set pattern parameters Pattern Disturbance Variables Pattern Template Disturbance Variables Pattern Parameters

α 1 ,..., α k – actor instances in the social network I(a i )=( α i1 ,…, α ir ) – instances of the actor a i S = (s 1 ,…,s t )= I(a 1 ) ×…× I(a n ) Pattern Language Algorithm for Pattern Application 1. Set pattern parameters 2. Instantiate disturbances Instantiate disturbances : D = ( d( s 1 ),…, d( s p )), where d( s i ) = d(v 1 / α i1 ,...,v n / α in , p 1 / μ 1 ,…, p m / μ m ) Pattern Disturbance Variables Pattern Template Disturbance Variables Pattern Parameters Pattern Template Instance Disturbance Instances Variables Pattern Parameters Digital Social Network

Pattern Language Algorithm for Pattern Application 1. Set pattern parameters 2. Instantiate disturbances 3. Evaluate disturbances 4a. Change Pattern Parameters 4b. Apply Pattern Solution Pattern Disturbance Variables Pattern Template Disturbance Variables Pattern Parameters Pattern Template Instance Pattern Instance Disturbance Variables Pattern Parameters Forces Force Relations Rationale Dependencies Description Solution Pattern Relations Disturbance Instances Variables Pattern Parameters Digital Social Network

ANT Subsystem Web Interface XML Repository Pattern Subsystem Formal Expression Module XML Pattern Repository Web Interface Social Network Subsystem Base Social Network Module JUNG Interface IBM DB2 Database Pattern Application Module Formal Expression Evaluation Pattern Instance Repository PALADIN Architecture Implementation PALADIN – PAttern LAnguage for DIsturbances in digital social Networks

Step 1 : define disturbance expression enter pattern properties Step 2 : bind variables to actors store pattern in the pattern repository PALADIN Web Interface

Troll Spammers Members Size reflects centrality of the member Members who participate in other disturbances, such as bursts or structural holes can be displayed as well PALADIN JUNG Interface Extension

Case study - 10 patterns of disturbance over 119 social network instances , 17359 individuals, 215 345 mails PALADIN Results Occurs in big networks where the members are distributed in different clusters. 40 No Leader Occurs for members having neighbours with only one contact. 67 Structural Hole Occurs in large networks where disconnected subnetworks exist. Scalability is necessary. 13 Independent Discussions The pattern occurs in the network centered around a member. 37 Leader Spammers can be found often in discussion groups. False positives exist. 86 Spammer Troll occurs very rarerly in cultural communities. True negatives exist. 2 Troll Occurs in small networks. The effects of the lack of an answering person must be further checked with content analysis. 61 No Answering Person The existence implies that the network is not popular. 67 No Questioner The existence implies little communication in the network. 76 No Conversationalist The pattern finds out topics which were very important for certain period of time. Scalability is necessary. 22 Burst Remarks Occurrences Pattern

Conclusion Depends on the used media in the network Relations built on the information from Google, FOAF, Mails, Bibliography Dependencies derived from the technical dependencies. Posting in the same thread. Relations Social Network Analysis, Semantic Web Individuals Friend-Of-A-Friend network, Google results Flink [Mika 2005] Disturbance-oriented, Pattern Repository, Social Network Analysis, Temporal Analysis, Statistics Media, Members, Artefacts Any Type of Digital Social Network PALADIN Temporal Analysis Developers, Software Components Eclipse IDE, CVS Repository Ariadne [de Souza et al. 2004] Social Network Analysis, Statistics Individuals, Mails, Threads, Genres Mailing List COMB [Boudourides et al. 2002] Analysis Approach Actors Media

Outlook Interoperability with applications based on Semantic Web, such as Flink Methodology for visualization of multidimensional disturbances, must reflect Media Artefacts SWAP-it [Seeling et al. 2004] InfoSky [Tochterman 2002] Dependencies Integration with simulation environment for social networks – can predict disturbances earlier

Multidimensional Patterns of Disturbance in Digital Social Networks

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Multidimensional Patterns of Disturbance in Digital Social Networks (20)

Recently uploaded (20)

Multidimensional Patterns of Disturbance in Digital Social Networks