Figure 2
From: Predicting speculation: a simple disambiguation approach to hedge detection in biomedical literature
Varying the number of non-zeros in the index vectors shows the effect of varying the number of non-zero elements in the random index vectors, while keeping the dimensionality fixed at k=5,000. The plot shows averaged and maximum sentence-level F1 across 5 different runs for each setting (using different random initializations of the index vectors), testing on 1/10th of the training data. For reference, the last column shows the result for using the original non-projected feature space.