Merging Temporal Annotations

In corpus linguistics obtaining high-quality semantically-annotated corpora is a fundamental goal. Various annotations of the same text can be obtained from automated systems, human annotators, or a combination of both. Obtaining, by manual means, a merged annotation from these, which improves the correctness of each individual annotation, is costly. We present automatic algorithms specifically for merging temporal annotations. These have been evaluated merging the annotations of three state-of-the-art systems on the gold standard corpora and the correctness of the merged annotation improved over that of individual annotations and baseline merging algorithms.