Annotating scientific publications with keywords and phrases is of great importance to searching, indexing, and cataloging such documents. Unlike previous studies that focused on user- centric annotation, this paper presents our investigation of various annotation characteristics on service-centric anno- tation. Using a large number of publicly available annotated scientific publica- tions, we characterized and compared the two different types of annotation processes. Furthermore, we developed an automatic approach of annotating scientific publications based on a machine learning algorithm and a set of novel features. When compared to other methods, our approach shows significant- ly improved performance. Experimental data sets and evaluation results are pub- licly availabl...

Click Here to View the Paper In XML Format