Scalable Corpus Annotation by Graph Construction and Label PropagationThomas Lansdall-Welfare, Ilias Flaounas, Nello Cristianini, Scalable Corpus Annotation by Graph Construction and Label Propagation. Proceedings of the 1st International Conference on Pattern Recognition Applications and Methods. ISBN 978-989-8425-98-0, pp. 25–34. February 2012. PDF, 559 Kbytes.
The efﬁcient annotation of documents in vast corpora calls for scalable methods of text classiﬁcation. Representing the documents in the form of graph vertices, rather than in the form of vectors in a bag of words space, allows for the necessary information to be pre-computed and stored. It also fundamentally changes the problem deﬁnition, from a content-based to a relation-based classiﬁcation problem. Efﬁciently creating a graph where nearby documents are likely to have the same annotation is the central task of this paper. We compare the effectiveness of various approaches to graph construction by building graphs of 800,000 vertices based on the Reuters corpus, showing that relation-based classiﬁcation is competitive with Support Vector Machines, which can be considered as state of the art. We further show that the combination of our relation-based approach and Support Vector Machines leads to an improvement over the methods individually.