CSE PhD Candidate Nathan LIU and Prof Qiang YANG Awarded Best Paper Award in the 20th ACM Conference on Information and Knowledge Management
CSE PhD candidate Nathan LIU and Prof Qiang YANG received the Springer Prize for Best Interdisciplinary Paper in the 20th ACM Conference on Information and Knowledge Management (ACM CIKM 2011) with their research "Transferring Topical Knowledge from Auxiliary Long Text for Short Text Clustering". The conference was held in Glasgow, UK in October 2011.
Since 1992, the ACM Conference on Information and Knowledge Management (CIKM) is one of the top conferences on topics in the general areas of databases, information retrieval, and knowledge management. Papers which bridge across these areas with special interest will be considered for the "Best Interdisciplinary Paper Award".
With the rapid growth of social Web applications, the task of understanding short texts is becoming more and more important. For short text messages, most of the existing text mining techniques are not effective due to the sparseness of text representations. The project shows that it is possible to find topically related long texts which can be utilized as the auxiliary data when mining the target short texts data. It presents a novel approach to cluster short text messages via transfer learning from auxiliary long text data. To accommodate the possible inconsistencies between source and target data, a novel topic model Du-al Latent Dirichlet Allocation (DLDA) model is proposed to jointly learn two sets of topics on short and long texts and couple the topic parameters to cope with the potential inconsistencies between data sets. Through large-scale clustering experiments on both advertisements and Twitter data, superior performance over several state-of-art techniques for clustering short text documents can be obtained.
900 submissions were received in the CIKM this year while only 134 (less than 15%) out of which were accepted as long papers with oral presentation. See the complete award list.