The Experts below are selected from a list of 12 Experts worldwide ranked by ideXlab platform
Christopher D Manning - One of the best experts on this subject based on the ideXlab platform.
-
a multi pass sieve for coreference resolution
Empirical Methods in Natural Language Processing, 2010Co-Authors: Karthik Raghunathan, Heeyoung Lee, Sudarshan Rangarajan, Nate Chambers, Mihai Surdeanu, Dan Jurafsky, Christopher D ManningAbstract:Most coreference resolution models determine if two mentions are coreferent using a single function over a set of constraints or features. This approach can lead to incorrect decisions as lower precision features often overwhelm the smaller number of high precision ones. To overcome this problem, we propose a simple coreference architecture based on a sieve that applies tiers of deterministic coreference models one at a time from highest to lowest precision. Each tier builds on the previous tier's Entity Cluster output. Further, our model propagates global information by sharing attributes (e.g., gender and number) across mentions in the same Cluster. This cautious sieve guarantees that stronger features are given precedence over weaker ones and that each decision is made using all of the information available at the time. The framework is highly modular: new coreference modules can be plugged in without any change to the other modules. In spite of its simplicity, our approach outperforms many state-of-the-art supervised and unsupervised models on several standard corpora. This suggests that sieve-based approaches could be applied to other NLP tasks.
Michael Fegan - One of the best experts on this subject based on the ideXlab platform.
-
partial sequencing of the hrpb and endoglucanase genes confirms and expands the known diversity within the ralstonia solanacearum species complex
Systematic and Applied Microbiology, 2000Co-Authors: Stephane Poussier, Jacques Luisetti, Chris Hayward, Philippe Prior, Michael FeganAbstract:We determined partial hrpB and endoglucanase genes sequences for 30 strains of Ralstonia solanacearum and one strain of the blood disease bacterium (BDB), a close relative of Ralstonia solanacearum. Sequence comparisons showed high levels of variability within these two regions of the genome involved in pathogenicity. Phylogenetic analysis based upon sequence comparisons of these two regions revealed three major Clusters comprising all Ralstonia solanacearium isolates, the BDB strain constituted a phylogenetically distinct Entity. Cluster 1 and Cluster 2 corresponded to the previously defined divisions 1 and 2 of Ralstonia solanacearum. Moreover, two subClusters could be identified within Cluster 2. The last Cluster, designated Cluster 3 in this study, included biovar 1 and N2 strains originating from Africa. This recently described group of strains was confirmed to be clearly different from the other strains suggesting a separate evolution from those of both divisions 1 and 2.
Philippe Prior - One of the best experts on this subject based on the ideXlab platform.
-
partial sequencing of the hrpb and endoglucanase genes confirms and expands the known diversity within the ralstonia solanacearum species complex
Systematic and Applied Microbiology, 2000Co-Authors: Stephane Poussier, Jacques Luisetti, Chris Hayward, Philippe Prior, Michael FeganAbstract:We determined partial hrpB and endoglucanase genes sequences for 30 strains of Ralstonia solanacearum and one strain of the blood disease bacterium (BDB), a close relative of Ralstonia solanacearum. Sequence comparisons showed high levels of variability within these two regions of the genome involved in pathogenicity. Phylogenetic analysis based upon sequence comparisons of these two regions revealed three major Clusters comprising all Ralstonia solanacearium isolates, the BDB strain constituted a phylogenetically distinct Entity. Cluster 1 and Cluster 2 corresponded to the previously defined divisions 1 and 2 of Ralstonia solanacearum. Moreover, two subClusters could be identified within Cluster 2. The last Cluster, designated Cluster 3 in this study, included biovar 1 and N2 strains originating from Africa. This recently described group of strains was confirmed to be clearly different from the other strains suggesting a separate evolution from those of both divisions 1 and 2.
Karthik Raghunathan - One of the best experts on this subject based on the ideXlab platform.
-
a multi pass sieve for coreference resolution
Empirical Methods in Natural Language Processing, 2010Co-Authors: Karthik Raghunathan, Heeyoung Lee, Sudarshan Rangarajan, Nate Chambers, Mihai Surdeanu, Dan Jurafsky, Christopher D ManningAbstract:Most coreference resolution models determine if two mentions are coreferent using a single function over a set of constraints or features. This approach can lead to incorrect decisions as lower precision features often overwhelm the smaller number of high precision ones. To overcome this problem, we propose a simple coreference architecture based on a sieve that applies tiers of deterministic coreference models one at a time from highest to lowest precision. Each tier builds on the previous tier's Entity Cluster output. Further, our model propagates global information by sharing attributes (e.g., gender and number) across mentions in the same Cluster. This cautious sieve guarantees that stronger features are given precedence over weaker ones and that each decision is made using all of the information available at the time. The framework is highly modular: new coreference modules can be plugged in without any change to the other modules. In spite of its simplicity, our approach outperforms many state-of-the-art supervised and unsupervised models on several standard corpora. This suggests that sieve-based approaches could be applied to other NLP tasks.
Stephane Poussier - One of the best experts on this subject based on the ideXlab platform.
-
partial sequencing of the hrpb and endoglucanase genes confirms and expands the known diversity within the ralstonia solanacearum species complex
Systematic and Applied Microbiology, 2000Co-Authors: Stephane Poussier, Jacques Luisetti, Chris Hayward, Philippe Prior, Michael FeganAbstract:We determined partial hrpB and endoglucanase genes sequences for 30 strains of Ralstonia solanacearum and one strain of the blood disease bacterium (BDB), a close relative of Ralstonia solanacearum. Sequence comparisons showed high levels of variability within these two regions of the genome involved in pathogenicity. Phylogenetic analysis based upon sequence comparisons of these two regions revealed three major Clusters comprising all Ralstonia solanacearium isolates, the BDB strain constituted a phylogenetically distinct Entity. Cluster 1 and Cluster 2 corresponded to the previously defined divisions 1 and 2 of Ralstonia solanacearum. Moreover, two subClusters could be identified within Cluster 2. The last Cluster, designated Cluster 3 in this study, included biovar 1 and N2 strains originating from Africa. This recently described group of strains was confirmed to be clearly different from the other strains suggesting a separate evolution from those of both divisions 1 and 2.