Workshop on Dynamic Networks Management and Mining (DyNetMM 2013)

SIGMOD/PODS Workshop on Dynamic Networks Management and Mining

In conjunction with SIGMOD/PODS 2013 

(New York, June 23, 2013)

Submission deadline: 22 FEBRUARY 2013

https://sites.google.com/site/dynetmm13/

===CFP===

The First ACM SIGMOD Workshop on Dynamic Networks Management and Mining
(DyNetMM 2013), to be held in New York, USA in conjunction with
SIGMOD/PODS 2013, is a venue for novel research on management and mining
in networks that change over time. The aim of DyNetMM is to disseminate
results founded on database indexing, access methods and mining that
advance the state-of-the-art in analysis, query processing, prediction and
mining in time-evolving networks arising in social-computational systems,
biology, transportation and others.

Both theoretical and systems/application papers are welcome to DyNetMM.
The workshop's focus is on methods that draw from techniques in databases,
indexing and data mining to enable scalable analysis, query support and
mining for patterns in dynamic relational data. The validation and novelty
of the submissions should match the high standards established in the
database community.

===Topics===

Submissions on the following general topics are welcome:
- Query languages and query support for dynamic networks
* Dynamic reachability queries
* Dynamic subgraph queries
* Predictive queries
- Indexing dynamic networks
* Neighborhood indexing
* Reachability and proximity indexing
- Tracking and prediction
* Road networks
* Geo-social graphs
- Mining in dynamic networks
* Frequent patterns
* Conserved and evolving patterns
* Anomaly and novelty detection
* Evolutionary clustering
- Network streams management
- Summarization and compression of dynamic networks
- Generative and predictive models of dynamic network evolution
- Distributed computation on dynamic graphs

===Organization===

Workshop co-chairs:
Petko Bogdanov, University of California Santa Barbara
Leman Akoglu , Stony Brook University
Ambuj K. Singh, University of California Santa Barbara
Christos Faloutsos, Carnegie Mellon University

===Program Committee===

Amol Deshpande (U Maryland)
Ananthram Swami (Army Research Lab)
Arun Maiya (IDA)
Boleslaw K. Szymanski (RPI)
Hanghang Tong (IBM)
Kathy Macropol ( Arcadia U)
Lance Kaplan (Army Research Lab)
Mayank Lahiri (Google)
Misael Mogiovi (UCSB)
Mohammed Zaki (RPI)
Prithwish Basu (BBN)
Rajmonda Caceres (MIT Lincoln Lab)
Ram Ramanathan (BBN)
Tina Eliassi-Rad (Rutgers)
U Kang (CMU)
Vagelis Papalexakis (CMU)
Veljko Pejovic (U Birmingham, UK)
Xifeng Yan (UCSB)

Read More

WWW 2013 Workshop - Mining Social network Dynamics (MSND 2013)

The 2nd International Workshop on 

Mining Social Network Dynamics (MSND 2013) 

To be held in conjunction with 

The 22nd International World Wide Web Conference (WWW) 

Rio de Janeiro, Brazil

May 13 - 17, 2013

CALL FOR PAPERS

In the real-time Web, the latest evolution of the Web, information is
generated very quickly, consumed by millions of users, and updated rapidly
by others through commenting, replying, transferring, etc. This is practiced
by people who differ in culture, knowledge, background, ideology. Moreover,
information generally comes from several channels and is sent out to
different ones. This is amplified by the social networking phenomenon, the
social Web, which is nowadays a well established set of technologies, based
on which users and service providers can exchange messages through an
interaction network, share information and collaborate, advertise a product,
create communities and influence them, etc. Besides, there is an abundant
literature regarding the different aspects of social networks be it their
construction or the detection of nodes playing specific roles.

However, it is well established that there is currently unclear
understanding of the laws governing these social networks, in particular in
the way they evolve over time. Thus, it is difficult to draw a clear image
linking the existing models of social networks and the real underlying
social mechanisms. As a result, there is a big gap in the evaluation and the
concretization of most of the research efforts in this area. Furthermore,
due to the growing complexity of digital social networks and the huge
quantity of new data available everyday, it becomes crucial for the
researchers to provide a clear understanding of the dynamics of these
networks. It also becomes important for the community to not only understand
what is happening currently in the network but also to predict the next
evolution and monitor the trends in the network. To efficiently analyze
these networks, it is important to be able to predict the dynamics in its
different forms: the content evolution (i.e., hot!
topics evolution), network structure (e.g., creation of new relations),
and information diffusion, influence evolution, etc.

Following the success of the last year's version of this workshop
(http://eric.univ-lyon2.fr/msnd/2012/), we are re-conducting it this year to
attract more researchers and make the event a central location where
researchers working on the issues of mining social network dynamics can meet
again to exchange ideas on open problems. We have also extended the topics
of interest for this workshop, with a particular interest towards dynamics.
And we aims at gathering researchers from the fields of social computing,
machine learning, and data mining to think about the obstacles that hurdle
the leveraging of understanding and capturing of social network dynamics. We
target researchers from both commercial and academic labs to join forces in
this exciting area. We intend to discuss the recent and significant
developments in the general area of mining social network dynamics and to
promote cross-fertilization of techniques. In particular, we aim at
identifying techniques from the data !
mining and machine learning fields that will enable researchers to
understand the dynamic phenomena in social networks and social media, as
well as specify important directions for the research communities.
Understanding, capturing, mining and being able to predict dynamic behaviors
is interesting for several areas such as marketing, security, and Web
search. To address the above mentioned aspects, we solicit the following
topics (but not limited to):

o Information diffusion in social networks;
o Community extraction, analysis, and evolution;
o Detection of (possibly evolving) roles;
o Content evolution and tracking in social networks;
o Social Journalism and news dynamics;
o Social networks affective and sentiment analysis;
o Social media recommendations;
o Information quality and evolution in social content;
o Security and privacy in rapidly evolving social networks;
o Evaluation techniques and benchmarks;
o New challenges in mining social networks;
o Example studies and use cases of dynamics of social
networks.

Paper submission
Submissions need to be formatted according to the ACM SIG Proceedings
Template. Two types of submissions are welcome: long papers with a limit of
8 pages or short ones with 4 pages length limit. We use the EasyChair system
for managing submissions, and the link to submit your paper is here:
http://eric.univ-lyon2.fr/msnd/. Further information will be communicated
soon regarding this special issue.

Key dates:

o Submission deadline: February 12, 2013
o Acceptance notification March 15, 2013
o Camera-Ready submission March 25, 2013
o Workshop date: May 13, 2013

Workshop Co-Chairs:
o Hakim Hacid, Bell Labs, France.
o Shengbo Guo, Xerox Research Centre Europe, France
o Athena Vakali, Aristotle University of Thessaloniki, Greece

Program committee (under construction)
o Julien Bourdaillet, Xerox, USA
o Yiannis Kompatsiaris, CERTH - ITI, Greece
o Amin Mantrach, Yahoo! Research Barcelona, Spain,
o Yosi Mass, IBM Research, Israel,
o Alessandra Sala, Bell Labs, Ireland
o Anna Stavrianou, Xerox Research Centre Europe, France
o Helen Paik, University of New South Wales, Australia
o Wei Peng, Xerox Research Centre Webster, USA
o Tetsuya Yoshida, Hokkaido University, Japan

Contact: msnd@eric.univ-lyon2.fr

Read More

PhD fellowships in network science, Central European University, Budapest, Hungary

Central European University Is Pleased To Announce Its Call For Applications
For Academic Year 2013/2014 PhD Certificate Program in Network Science

Central European University is an English-language, graduate university located in Budapest, Hungary. An
international university with students from some 100 countries and faculty from more than 40, CEU offers small, seminar-style courses and a student/faculty ratio of 7:1. CEU is accredited in the United States and Hungary.

The PhD Certificate Program in Network Science is a new program, offered by the Center for Network Science, that provides tools to understand the principles governing social, political, economic, and environmental networks. Students get to work with some of the most prominent network scientists in the world (Albert-Laszlo Barabasi, Rosario Mantegna, Janos Kertesz), exploring terabyte-sized datasets that capture mobile phone calls between millions of individuals; financial transactions between banks and companies; energy delivery systems; memberships in the directorial boards of companies; and affiliations in political parties.

Candidates with an interest in network science and its applications to political science, economics, and
environmental sciences and policy can apply via the respective departments' doctoral programs. Upon
completion of the doctoral program, students will also receive a PhD certificate in network science.

The program focuses on the following problem areas:
• Political science: business-political networks, social movements, social influences in voter behavior
• Economics: world system networks and development, embeddedness of multinational corporations,
network effects and externalities
• Environmental science: resilience in energy grids, food webs and economic competition

Scholarships and Fellowships

CEU is committed to attracting talented students and scholars from around the world. The University provides a variety of scholarships and research grants for which applicants from any country are eligible to apply.
• CEU Doctoral Fellowships cover full tuition, health insurance, and a generous stipend for housing and living
expenses. More than 90 percent of PhD students receive the full doctoral fellowship from CEU, while others
are supported by external organizations providing funding.
• Direct Student Loans and Canada Student Loans are available for eligible U.S. and Canadian students.
• External Scholarships and Financial Aid Programs administered through CEU include Erasmus Mundus
scholarships, Erasmus research grants, and other private and public fellowships and scholarships.

Career Paths

Network scientists are employed in academia, government, and business. Applied network scientists work
at Internet and social media companies, where they improve users' online experience. They also serve as
consultants developing business infrastructure and specialize in risk mitigation for major financial institutions.
Application Deadline: January 24, 2013
More Information: http://tinyurl.com/c4ys646

Read More

Dynamic Network Analysis for Computer Science Publications

Dynamic network properties of ICALT, ICWL and ECTEL

AERCS is a visualization tool for communities of academic events in Computer Science. It is developed and hosted by the Advanced Community Information Systems (ACIS) group at  RWTH Aachen University, Germany. Currently, conferences and journals from the DBLP dataset are visualized. Other data sets will be integrated in the future.  While this bibliographic data set is quite comprehensive it does not show all publications!

Dynamic network analysis uses time series analysis of networks to explore dynamic patterns of networks such as growth and decay. A typical result of our research is the dynamic comparison of computer science communities. A computer science conference or journal can be compared by time series analysis for typical social network growth patterns like density, diameter, clustering coefficient, largest connected component, average path length and maximum betweenness. In our example above, we compare three major conferences in technology enhanced learning: the EC-TEL, the ICALT and the ICWL community. Here, we visualize the time series analysis of the co-authorship network of the three conferences.
For the co-authorship network, the emergence of the giant component (largest component) indicates the cohesiveness of collaboration within the community, while the betweenness shows the existence of gatekeepers and their importance. The clustering coefficient measures the extent to which the community is clustered into sub-communities. Other parameters such as diameter and average shortest path length, show whether the community is still developing or whether it is stable. A complete analysis of the three communities will be published in a future issue of the Journal of Educational Technology & Society.

Everybody can use this service for comparing journals and conference in computer science by means of dynamic network analysis. Just visit AERCS and test it out!


This service is brought to you by the chair of information systems and databases at RWTH Aachen University.

Read More