:irwinking07-1.jpg

Introduction to Social Computing and Its Computation Techniques

Irwin King, Chao Zhou, and Baichuan Li

Department of Computer Science and Engineering
The Chinese University of Hong Kong

Abstract

With the emergence of Web 2.0, social networks have becoming an integral and important part of our changing social cultural. With the novel transformative ways to connect, collaborate, and create communities on the Web, the phenomenon of cyber social behaviors have emerged that intrigue researchers and practitioners alike. Currently, we have voluminous data collected from web sites, blogs, wikis, clickthrough data, query logs, tags, etc.~from areas such as social networks, social search, social media, social bookmarks, social news, social knowledge sharing, and social games. These data provide a wealth of information available for us to process, analyze, and mine.

Social Computing involves the investigation of collective intelligence by using computational techniques such as machine learning, data mining, natural language processing, etc. The key characteristic of Social Computing that is different from other type of intelligent computation is the availability of (1) relevant personalized information and (2) the social network, i.e., relational information, among the users. Without these two sources of information, many applications can simply use existing theories, models, algorithms, and applications for processing the information. However, social computing is about how to use these two rich sources of social information to effectively and efficiently compute interesting results.

There are several important components in the investigation of social computing. First, a better formal theory and model about cyber social interactions would be important so that future social interactions and phenomenon can be estimated and/or predicted. Second, better algorithms to mine existing spatial (relational) and temporal (time varying) data with efficiency would be needed. In particular, ways to deal with partial information and incomplete information in systems such as recommender systems, tagging systems, etc.~will be important to ensure the computed results are appropriately accurate and adequate. Third, some may examine these algorithmic issues from the scalability view point. Since social networks may involve with complex individual and community relationships, algorithms for computing any functional results must be efficient and scalable to cater to an ever increasing Web. Fourth, security and privacy issues are of grave concern on the Web, specially in social networks. Theories and algorithms to protect important personal information is important when relations are easily created and difficult to eliminate. Moreover, ways to sanitize data for research have also attracted much attention lately due to the importance of conducting social computing research work, e.g., query log, clickthrough data, etc. Lastly, one interesting and hotly discussed issue is the monetization of social interaction/computing. Here, the matter turns to finding ways for making financial gains from social computing. Although this may be interesting, we plan not to discuss this topic in depth.

The tutorial will conclude by summarizing and reflecting back on the cyber social behavior trends that we are observing on the Web and posit that what we have presented in the tutorial is just a tip of the iceberg to a whole area of exciting and dynamic research that is worthy of more detailed investigation for many years to come.

Research Interests

Irwin King's research interests include social computing, machine learning, web intelligence, and multimedia processing. In these research areas, he has over 200 technical publications in journals, conferences, book chapters, and edited volumes. He was the General Chair of Web Search and Data Mining (WSDM2011), and also has been involved with the organization and/or technical program of many international conferences such as WWW, SIGIR, KDD, AAAI, etc. Moreover, he has served as reviewer and panel member for Research Grants Council (RGC) of Hong Kong, Natural Sciences and Engineering Research Council of Canada (NSERC), National Natural Science Foundation of China (NSFC), and Natural Science, and Engineering of Academy of Finland. Dr. King is an Associate Editor of ACM Transactions on Knowledge Discovery from Data (ACM TKDD) and a former Associate Editor of the IEEE Transactions on Neural Networks (TNN). He is a member of the Editorial Board and Special Issue Editor of a number of international journals. He is Professor at the Department of Computer Science and Engineering, The Chinese University of Hong Kong. Currently he is on leave with AT&T Labs Research, San Francisco and is also teaching Social Computing and Data Mining as a Visiting Professor at UC Berkeley. He received his B.Sc. degree in Engineering and Applied Science from California Institute of Technology, Pasadena and both his M.Sc. and Ph.D. degrees in Computer Science from the University of Southern California, Los Angeles. See http://irwinking.com for more information.

Presentation Materials

  1. Introduction: pdf
  2. Matrix factorization: pdf
  3. Link analysis and question/answering: pdf
 
confs/wcci2012-tutorial-socialcomputing.txt · Last modified: 2012/06/09 23:53 by admin     Back to top