The main goal of this thesis is to discover the high-quality communities for academic papers. Since
tremendous amounts of academic papers have been published, many researchers experience diculties in
exactly finding the papers in which they are interested. Community detection (or cluster discovery) can
facilitate this task since it finds similar or relevant papers. Thus, community detection from academic
papers has received a lot of attention, but it is still a challenging issue. Along this direction, citation
analysis and attribute (e.g., content) analysis have been widely used. However, most existing methods
focus on either citation analysis or attribute analysis, disregarding the other side. The novelty of this
thesis is a complete merger between citation analysis and attribute analysis. Our approach constructs
a network of academic papers by considering both types of information together and then performs
clustering to obtain the communities of papers. In the network, an edge between two papers is created
by considering both (i) the existence and importance of citations from one to the other and (ii) the
attribute similarity between the two papers. In this way, the two types of information are considered at
the same time, not sequentially. The optimal merger was empirically determined. Last, the effectiveness
of our approach was verified by extensive experiments. About half-million papers were crawled, and the
full text was extracted from them for attribute analysis. The results show that our approach produces
higher-quality communities compared with the baseline approaches that use either citation analysis
or attribute analysis. Overall, we believe that our approach will be very useful for academic search
engines.