| Experimental Setup | We run four different clusterings for each base set size (except for the large sets, see below). |
| Experimental Setup | The unique-event clusterings are motivated by the fact that in the Dupont—Rosenfeld model, frequent events are handled by discounted ML estimates. |
| Experimental Setup | As we will see below, rare-event clusterings perform better than all-event clusterings . |
| Results | When comparing all-event and unique-event clusterings , a clear tendency is apparent. |
| Clustering phrase pairs directly using the K-means algorithm | Using multiple word clusterings simultaneously, each based on a different number of classes, could turn this global, hard tradeoff into a local, soft one, informed by the number of phrase pair instances available for a given granularity. |
| Clustering phrase pairs directly using the K-means algorithm | In the same fashion, we can incorporate multiple tagging schemes (e.g., word clusterings of different gran-ularities) into the same feature vector. |
| Experiments | Figure 1 (left) shows the performance of the distributional clustering model ( ‘Clust’ ) and its morphology-sensitive extension (‘Clust—morph’) according to this score for varying values of N = l, . |
| Experiments | , 36 (the number Penn treebank POS tags, used for the ‘POS’ models, is 36).6 For ‘Clust’ , we see a comfortably wide plateau of nearly-identical scores from N = 7,. . |