October 2008 Archives

Dr. Mark Newman is very famous in social network analysis. He has published and edited a lot of influential books and papers in this area, which can be found here. His research field is statistical physics theory, and his research focus on networks, spin systems and percolation, Monte Carlo methods. I cannot say he is the pioneer of social networks, but he is the pioneer during the recent great development of social networks, especially in the community structure discovery direction.

He got his B.A. in University of Oxford, 1988 and then he received his Ph.D. still in University of Oxford in 1991. Right now he is the Paul Dirac Collegiate Professor of Physics and member of the external faculty of Santa Fe Institute.

Dr. Newman studies the structure and function of networks, including social and biological networks and computer networks like scientific coauthorship networks, citation networks, email networks, friendship networks, epidemiological contact networks, and animal social networks. He employs a combination of empirical methods, analysis, and computer simulation during the research work. And his research of social networks focuses on several directions as follows:

(1)    He studies some fundamental network attributes, such as degree distributions, centrality measures, assortative mixing, vertex similarity.

(2)    His work of social networks is the most famous known for the study of community structure. He proposed Girvan-Newman algorithm in 2002, which is one of the most important and classic algorithm in community detection. His research in this area involves how contact networks form and how structures affect the diffusion of information. Also, he has carried out empirical work on the collaboration networks, and developed computer simulations of the growth and formation of networks.

(3)    Recently, he works on models of information propagation like disease propagation, friendship formation and spread of computer viruses over email networks.

(4)    His other recent projects include navigation in networks and the Internet, mixing patterns within networks, network correlations and phase transitions in network structure.

Last time, I briefly introduced some academic communities related to Social Network Analysis.  In this blog, I would like to describe several academic publication or presentation venues that I hope to have my work in someday.

1. WWW: As I mentioned in the previous article, WWW is the leading conference in the web field which focuses on the future development of the World-Wide Web and its impacts to the world. In the recent years, Social Network has been discussed a lot in this conference.

WWW2009 will be held in Madrid, Spain on April 20-24, 2009. The deadline of paper is November 03, 2008 and the notification date is January 20, 2009. WWW2009 will include the track of Social Networks and Web 2.0.

2. IJCAI: IJCAI is a top conference in the AI area which both surveys the mature area of AI research and/ or practice and introduces novices to major topics of AI. Its impact factor is 1:1.82 (top 4.09%). It is hard to publish in IJCAI not just because that it has a low rate of acceptance, but also it is held every two years. Since Social Network is a branch of AI, it is reasonable to watch this conference closely.

IJCAI2009 will be held in Pasadena, California, on July 11-17, 2009. The deadline of abstract is January 7, 2009, the deadline of paper is January 12, 2009 and the notification date is March 31, 2009.

3. AAAI: AAAI is also a top conference in the AI field as IJCAI. It devotes to advancing the scientific understanding intelligent thought and behavior. Its impact factor is 1:1.87 (top 3.60%).

AAAI 2009 Spring Symposia will be held at Stanford University on March 23-25, 2009. The deadline of paper is October 03, 2008 and the notification date is November 07, 2008. 

4. SIGKDD: SIGKDD is the most important conference in data mining. The primary focus of SIGKDD is to discuss the development in knowledge discovery and data mining, and the adoption of standards in this area. Its impact factor is 1:1.68 (top 6.14%). This conference has a lot of discussions related to social network, for the reason that social network can either use data mining methods and can be applied to it.

SIGKDD-09 will be held in Paris, France on June 28 to July 01, 2009. The deadline of abstract is February 2, 2009, the deadline of paper is February 6, 2009 and the notification date is February 13, 2009.

Right now I am working on Social Network Analysis, which is related to Web and Information Systems (W3), Artificial Intelligence (AI), Data Mining (DM) and Databases (DB).

For the Web and Information Systems area, the World-Wide Web Conference (WWW) is the leading conference community, which focuses on the future evolution of the Web, the standards and the impacts of the Web to the society. The topics of WWW include Browsers and UI, Data Mining, Mobility, Multimedia, Search, Security and Privacy, Semantic Web, Social Networks and Web 2.0 and so on. Social Network is recently been discussed in WWW and has just developed into workshops - Workshop on Social Web and Knowledge Management (SWKM 2008) and Workshop on Social Web Search and Mining (SWSM 2008). From the workshops, I can learn what research about social network is being conducted and then I can get the research trends, which are very helpful for my research work.

In the Artificial Intelligence field, there are several important and influential conferences, such as International Joint Conference on Artificial Intelligence (IJCAI), American Association for AI National Conference (AAAI), International Conference on Machine Learning (ICML) and Neural Information Processing Systems (NIPS). Generally speaking, the first two conferences discuss different topics of AI while the last two focus on specific topics - machine learning and neural information separately. Since social network is a branch of AI, a lot of influential papers of social network can be found in these conference communities.

Also, social network can either use the data mining and databases methods and can be applied in these two areas. In the data mining area, ACM Knowledge Discovery and Data Mining (SIGKDD) and IEEE International Conference on Data Mining (ICDM) are the top tier conferences. And in the databases field, Very Large Data Bases (VLDB), ACM SIGMOD Conference on Management of Data (SIGMOD) and IEEE International Conference on Data Engineering (ICDE) are the leading conference communities. Also, International Conference on Information and Knowledge Management (CIKM) is also important in database communities. During the conferences of DM and DB listed above, a lot of interesting topics and discussions of social network can be found.

Shizhuo Zhu is the only senior IST student in our lab - the Laboratory for Intelligent Agents, though there are several seniors in the department of Computer Science and Engineering. Shizhuo has been in IST since 2003, and right now he is preparing for his Ph.D. dissertation defense. The topic of his dissertation is "Hypothesis-Driven Story Building: A Framework for Supporting Decision-Making as Partial Information Arrives Over Time", which focuses on the attributes of decision-making especially when information is incomplete and changing.

Shizhuo's research interests mainly focus on artificial intelligence and multi-agent systems. And he has attended several conferences as follows:

1. International Human-Computer Interaction Conference (HCI), Crete, Greece, 2003.

2. North East Student Colloquium on Artificial Intelligence (NESCAI), Ithaca, USA, 2006.

3. Conference on Artificial Intelligence in Medicine (AIME), Amsterdam, Netherlands, 2007.

And he has published 15 papers including the ones he published when he was a master student in China, such as:

1. Shizhuo Zhu, Joanna Abraham, Sharoda A. Paul, Madhu Reddy, John Yen, Mark Pfaff, and Christopher DeFlitch, R-CAST-MED: Applying Intelligent Agents to Support Emergency Medical Decision Making Teams. The 11th Conference on Artificial Intelligence in Medicine (AIME'07), Amsterdam, Netherlands, July 09-11, 2007.

2.  Po-Chun Chen, Xiaocong Fan, Shizhuo Zhu, John Yen: Boosting-Based Learning Agents for Experience Classification. In Proceedings of the 2006 IEEE/WIC/ACM International Conference on Intelligent Agent Technology (IAT'06), p. 385-388, 2006.

3. Yanqing Ji, Hao Ying, John Yen, Shizhuo Zhu, Daniel C. Barth-Jones, Richard E. Miller, R. Michael Massanari, A Distributed Adverse Drug Reaction Detection System Using Intelligent Agents with Fuzzy Recognition-Primed Decision Model, International Journal of Intelligent Systems, in press.  

A complete list of his publication can be found at http://my.win.psu.edu/szz104/pulications.htm. 

Academically, he thinks himself focus more on technology though he is an IST student. During his research work, he enjoys the happiness from new discoveries while most time he has to bear the loneliness and depression. Maybe this is what Ph.D. students do, in order to find new research discoveries, we have to pay a lot of time and attention and most time we are destined to be painful and lonely. But who knows, maybe I will find the research is worth it.

Both of Shizhuo's and my research work lie on Artificial Intelligence (AI), but he focus on the direction of intelligent agents like decision-making while mine is social network analysis. Though the two research topics above look very different, we may use a lot of same and similar AI thoughts and methods during our research work. 

Note: More details about Shizhuo Zhu can be found at his webpage: http://my.win.psu.edu/szz104/.

Dr. John Yen has done a lot of research work in the AI area and he has published more than 100 papers. He is currently the Vice President of Publication for IEEE Computational Intelligence Society, the Chair of IEEE FIPA Working Group on Human Agent Communications, a Sponsoring co-Chair of AAMAS 2008, and a member of ACM Senior Member Committee.

Dr. John Yen got his bachelor's degree in Electrical Engineering from Honors National Taiwan University in 1980. Then he came to USA for his graduate study, he got his master's degree in Computer Science from University of Santa Clara in 1982. After that, he came to University of California, Berkely and got his doctor's degree in Computer Science in 1986. There, he met his advisor, Lofti Zadeh, who is the father of fuzzy logic.

After graduation, Dr. John Yen spent three years at USC Information Sciences Institute (ISI) and did research work on logic-based knowledge representation scheme with production rules. Then, he joined Texas A&M University and worked on fuzzy logic, robotics and intelligent systems. In 2001, he attended College of IST, Penn State University.

Dr. John Yen has broad research interests, which mainly focus on agents and network analysis. He is interested in research that transforms information into knowledge, decisions and actions, which can facilitate human teams. Also, he works on extracting knowledge from large-scale social networks and modeling the dynamic growth of them.