Thursday, April 15, 2010

ISiM Workshop on SNA

A three day workshop on the social network analysis was organized at the ‘International School of Information Management (ISiM)’, University of Mysore. This document elaborates the details of the 3 day event.

About the workshop/relevance:
Social network analysis as a field of study is growing rapidly and in popularity. It is now evolving into a new paradigm across academia, business, industry, popular culture and folklore. It is both an approach and a tool to uncover and understand the hidden side of connections that drive certain phenomenon involving a network of human players. SNA is the technique of mapping and monitoring the relationships and flow of knowledge between individuals, teams, organizations, electronic devices, URLs and other interconnected entities. With the success of social networking tools like Facebook, Twitter, Myspace and Orkut with the masses, their influence on the group behavior and opinions are being increasingly felt every day. SNA is since being perceived from being a suggestive metaphor to an analytic approach to a paradigm, with its own theoretical statements, methods, social network analysis software, and researchers.

Workshop Objective:
Introduce the audience to the goals and perspectives of network analysis.
o Understanding of network data and issues related to collection, validity, visualization and mathematical/computer representation.
o Methods of detection and description of structural properties such as centrality, cohesion, subgroups, cores, roles etc.

At the end of the course, the participants will be,
o Able to examine data in 'social networks way'
o Identify and formulate network analysis problems
o Solve them using network analysis software and Interpret the obtained results.

About the Faculty:

Kevin Crowston is a Professor of Information Studies at the Syracuse University School of Information Studies since 1996. He received his A.B. (1984) in Applied Mathematics (Computer Science) from Harvard University and a Ph.D. (1991) in Information Technologies from the Sloan School of Management, Massachusetts Institute of Technology (MIT). His current research interests include: empirical studies of coordination-intensive processes in human organizations; theoretical characterizations of coordination problems and alternative methods for managing them; and design and empirical evaluation of new kinds of computer systems to support people working together.

Facilities at the workshop:
Accommodation was offered to the outstation participants at a discounted rate at ‘The Fern Residency’. The delegates were handed the workshop kit that included A bag pack, Workshop presentation Material, Workshop CD (Pajek software, Presentation slides, Reference book etc), Notepad and a pen. The participants who were not carrying their own laptops were provided with individual desktops/laptops to be used during the workshop.

Workshop Schedule:
The contents of the workshop were spread over 3 days. The detailed session plan was as mentioned below.

Workshop Participants :
There were 29 participants in all for the workshop. They were drawn from all parts of the country and abroad. The classification of the participants is pictorially represented below.

Day 1 Proceedings:
The workshop started with the introduction of the participants. Each of them explained their objectives behind attending the workshop. The workshop started with running through the basics of graph theory. The distinction between the regular network analysis and the social network analysis was brought out. The participants were then informed about the various social networking relations that can exist between individuals – friendship, Liking, business transactions, information/knowledge sharing, formal relations, biological relations, associations and affiliations being few of them.

After the tea break, issues related to network analysis were taken up. Network was defined as the sum of the graph and its constituent data. The graph itself is constituted of two elements – Nodes and Links. Therefore network can be redefined as the data that resides on the nodes and that which can be interpreted by the links. More concepts were understood such as the size of the network, density of the graph, Average degree, directed network, reciprocity of the network, adjacency matrix, collecting network data and completeness of data.
After the lunch session, the participants were introduced to Pajek – Large Network Analysis tool. The ‘World Trade’ dataset was taken as an example to understand how the tool can be leveraged to understand the network and the social interactions among the nodes.

Day 2 Proceedings:
The day started with the review of previous day’s sessions. ‘Knoke’s Exchange Network’ (KnokBur) was taken as an example to understand some of the concepts of power law networks and degree distribution. After this, the issue of ‘Reachability’ was taken up for discussions. The concepts like ‘Walks’, ‘Paths’, ‘Cycles’, ‘Geodesic distance’ and ‘diameter’ were discussed as applicable to the social network analysis.
More concepts were introduced such as structural holes and Brokerage. These concepts help in identifying the centrality (nodes which are the most important in the network). It was learnt that even the central nodes can have varying degrees of centralization. Prof. Crowston explained further that based on the number of links that a node receives compared to the links that it sends, the ranking of a node can be determined. He also explained how different kinds of networks have to be analyzed differently to assign appropriate rankings to its nodes.

After the lunch break, the participants worked on various datasets/cases to implement their learnings. Some of them even tried to extract fresh network data for the purpose.

Day 3 Proceedings:
The final day’s session mainly concentrated on the concepts of ‘Position and Social Roles’. Positioning refers to identifying the nodes across networks that have similar attributes as defined by the patterns of relations. Social roles refer to the relation and identification of one node with respect to the other. For example, a ‘father’ is defined by the mandatory presence of the ‘child’ node etc.

During the lunch break, the team gathered for a group photograph. And after the lunch session, participants worked with datasets to explore more network concepts using the Pajek software. The workshop concluded with Prof. Kevin Crowston handing over the certificates to the participants.

Pictures taken during the event 

Day 1

Day 2

Day 3

To know more, please write to Mr. Jayanth Jagadeesh at ( /

No comments:

Post a Comment