We have seen the explosive growth of social network and it provides a huge amount of data. We have millions of tweets about the brands, thousands of Facebook “likes,” hundreds of thousands of check-ins on Foursquare. So what are we going to do with this large data? How to construct the huge data fragment? What underlying knowledge can we explore from this huge data? Thus, we derive the problem of user identity mapping across social network. Basically, The key challenge is to construct a discriminative social signature based on a variety of elements in different social network.
Members
The identification of a user’s core community benefits in the following aspects. First, it is observed on Twitter that a user’s online social network, i.e., the part of the follow network excluding the core community, is more informative of the user’s interest, hobby, life-style, etc. Second, the core community, on the other hand, contributes to a more robust and accurate profiling of the user’s interests. This is because one’s closest friends in real life are likely to be of a similar kind. Finally, the discovery of the core community gives insight into the different characteristics of a user’s online and off-line social network, which is interesting in itself.
Members
Gaming is an essential part of human activity that brings enjoyment to players. To derive pleasure from gaming, a player has to be able to gain sufficient expertise to achieve the desired outcomes. This gaming expertise is usually accumulated through playing or watching many game instances. Because of the recent advances in database systems and collaborative web sites, it is possible to record massive amount of game data generated by player. One can then apply data analytics to discover game strategies, player performance, and important behavior patterns.
Members
Pinnacle Lab