REDI (2017-2B) Research Experiments in Databases and Information Retrieval

Welcome to REDI! Federated social networks may be the answer to scandals with, and abuse of, personal data by large social networks such as Facebook. In this course, students will research how recommendations (such as "Who to follow?" or "Trending topics") can be implemented in federated social networks. A challenge when implementing such recommendations is that in federated social networks, a recommender system cannot use nor know all data in the network.

Students will implement and evaluate a bot on Mastodon that sends its followers (personal) recommendations. Students learn answers to questions like: How do centralized social networks implement recommenders? How to build recommenders with incomplete data? How to evaluate a recommender with user-interaction data? How to write a scientific paper?

Follow #REDI on Mastodon for course updates.

Wishing you a fruitful course,    Djoerd Hiemstra.

Reading Material

• Jean-luc Doumont,  ed. English Communication for Scientists Unit 2, Writing Scientific Papers. Cambridge, MA: Nature Education, 2010
• Richard Esguerra. An Introduction to the Federated Social Network, Electronic Frontier Foundation (EFF), 2011
• Christopher Lemmer Webber, Jessica Tallon, Erin Shepherd, Amy Guy, Evan Prodromou. ActivityPub W3C Recommendation, World Wide Web Consortium,  2018
• Pankaj Gupta, Ashish Goel, Jimmy Lin, Aneesh Sharma, Dong Wang, Reza Zadeh. WTF: The Who to Follow Service at Twitter. World Wide Web Conference, 2013 (presented by Group 6)
• Luca Maria Aiello, Georgios Petkos, Carlos Martin, David Corney, Symeon Papadopoulos, Ryan Skraba, Ayse Goker, Yiannis Kompatsiaris, Alejandro Jaimes. Sensing trending topics in Twitter, IEEE Trans Multimed 15: 1268–1282, 2013 (presented by Group 1)
• Jure Leskovec and Christos Faloutsos. Sampling from Large Graphs. Proceedings of SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2006. (presented by Group 4)
• M. Brautbar and M. Kearns. Local algorithms for finding interesting individuals in large networks. Proceeding on the Innovations in Computer Science (ICS), pages 188–199, 2010 (presented by Group 3)
• Christian Borgs, Michael Brautbar, Jennifer Chayes, Sanjeev Khanna, Brendan Lucier. The power of local information in social networks. International Workshop on Internet and Network Economics (WINE) Internet and Network Economics, pp 406-419, 2012.
• K. Avrachenkov, Nelly Litvak, L. Ostroumova Prokhorenkova, E. Suyargulova. Quick detection of high-degree entities in large directed networks. Proceedings of the IEEE International Conference on Data Mining (ICDM), 2014.
• Thorsten Joachims. Evaluating Retrieval Performance Using Clickthrough Data. Proroceedings of the ACM Conference on Knowledge Discovery and Data Mining (KDD), 2002.
• Filip Radlinski, Madhu Kurup, and Torsten Joachims. How does clickthrough data reflect retrieval quality?  In Proceedings of the 17th ACM conference on Information and knowledge management (CIKM), pages 43–52, 2008.

Additional reading material

• Tianyi Wang, Yang Chen, Zengbin Zhang, Tianyin Xu, Long Jin, Pan Hui, Beixing Deng, Xing Li. Understanding Graph Sampling Algorithms for Social Network Analysis, Proceedings of the 31st International Conference on Distributed Computing Systems Workshops, 2011 (presented by Group 2)
• Ruturaj Dhekane and Brion Vibber. Talash: Friend Finding In Federated Social Networks. WWW Workshop on Linked Data on the Web (LDOW), 2011 (presented by Group 5)
• Panagiotis Symeonidis and Christos Perentis. Link Prediction in Multi-modal Social Networks. Joint European Conference on Machine Learning and Knowledge Discovery in Databases (ECML PKDD), 2014 (presented by Group 8)