Tree Kernels for Semantic Role Labeling

By Alessandro Moschitti, Daniele Pighin, Roberto Basili. 2008

179 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Extraction

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Aliaksei Severyn
2. Alessandro Moschitti
3. Regina Barzilay
4. Tao Lei
5. Danilo Croce
6. Roberto Basili
7. Lluís Màrquez i Villodre
8. Bonaventura Coppola

In Information Extraction

1. Aliaksei Severyn
2. Alessandro Moschitti
3. Bonaventura Coppola
4. Danilo Croce
5. Roberto Basili
6. Yuan Zhang
7. Regina Barzilay
8. Lluís Màrquez i Villodre

Citations in the corpus

• A General Purpose FrameNet-based Shallow Semantic Parser (2010, methodology): “Thus, to achieve independence on the application domain, we exploited Tree Kernels (Collins and Duffy, 2002) over automatic structural features proposed in (Moschitti et al., 2005a; Moschitti et al., 2008).”
• Using Sequence Kernels to identify Opinion Entities in Urdu (2011, background): “English has seen extensive use of sequence kernels (string and tree kernels) for tasks such as relation extraction (Culotta and Sorensen, 2004) and semantic role labeling (Moschitti et al., 2008).”
• Auditing Deep Learning processes through Kernel-based Explanatory Models (2019, background): “As proposed in (Moschitti et al., 2008), SRL can be modeled as a multi-classification task over each parse tree node n , where argument spans re-flect sub-sentences covered by the tree rooted at n .”
• Structured Lexical Similarity via Convolution Kernels on Dependency Trees (2011, methodology): “This is very valuable as previous work showed that tree kernels (TK) alone perform lower than models based on manually engineered features for SRL tasks, e.g., (Mos-chitti, 2004; Giuglea and Moschitti, 2004; Giu-glea and Moschitti, 2006; Moschitti, 2006b; Che et al., 2006; Moschitti et al., 2008).”
• High-Order Low-Rank Tensors for Semantic Role Labeling (2015, methodology): “Other approaches have focused on semantic generalizations of lexical features using selectional preferences, neural network embeddings or latent word language models (Zapirain et al., 2013; Collobert et al., 2011; Deschacht and Moens, 2009; Roth and Woodsend, 2014).”
• Modelling input texts: from Tree Kernels to Deep Learning (2015, methodology): “For this purpose, we train a binary Boundary Classifier using the AST subtree defined in [97], i.”