Facing the most difficult case of Semantic Role Labeling: A collaboration of word embeddings and co-training
نویسندگان
چکیده
We present a successful collaboration of word embeddings and co-training to tackle in the most difficult test case of semantic role labeling: predicting out-of-domain and unseen semantic frames. Despite the fact that co-training is a successful traditional semi-supervised method, its application in SRL is very limited. In this work, co-training is used together with word embeddings to improve the performance of a system trained on CoNLL 2009 training dataset. We also introduce a semantic role labeling system with a simple learning architecture and effective inference that is easily adaptable to semi-supervised settings with new training data and/or new features. On the out-of-domain testing set of the standard benchmark CoNLL 2009 data our simple approach achieves high performance and improves state-of-the-art results.
منابع مشابه
The Semantics of the Word Istikbar (Arrogance) in the Holy Quran based on Syntagmatic Relations(A Case Study of Semantic Proximity and Semantic Contrast)
The word istikbar (arrogance) is one of the key words in the monotheistic system of the Quran, which has found a special status as a special feature of the opponents and adversaries of the call to the truth. Given the prominent role of this issue in the human life system and its provision of corruption and moral deviations, it is necessary to represent the nature of the elements that make up th...
متن کاملبرچسبزنی نقش معنایی جملات فارسی با رویکرد یادگیری مبتنی بر حافظه
Abstract Extracting semantic roles is one of the major steps in representing text meaning. It refers to finding the semantic relations between a predicate and syntactic constituents in a sentence. In this paper we present a semantic role labeling system for Persian, using memory-based learning model and standard features. Our proposed system implements a two-phase architecture to first identify...
متن کاملLabeling Subgraph Embeddings and Cordiality of Graphs
Let $G$ be a graph with vertex set $V(G)$ and edge set $E(G)$, a vertex labeling $f : V(G)rightarrow mathbb{Z}_2$ induces an edge labeling $ f^{+} : E(G)rightarrow mathbb{Z}_2$ defined by $f^{+}(xy) = f(x) + f(y)$, for each edge $ xyin E(G)$. For each $i in mathbb{Z}_2$, let $ v_{f}(i)=|{u in V(G) : f(u) = i}|$ and $e_{f^+}(i)=|{xyin E(G) : f^{+}(xy) = i}|$. A vertex labeling $f$ of a graph $G...
متن کاملWord clustering effect on vocabulary learning of EFL learners: A case of semantic versus phonological clustering
The aim of this study is to determine the effect of word clustering method on vocabulary learning of Iranian EFL learners through a case of semantic versus phonological clustering. To this effect, 80 homogeneous students from four intermediate classes at an English institute in Torbat e Heydariyeh participated in this research. They were assigned to four groups according to semantic versus phon...
متن کاملDistributed Representations for Unsupervised Semantic Role Labeling
We present a new approach for unsupervised semantic role labeling that leverages distributed representations. We induce embeddings to represent a predicate, its arguments and their complex interdependence. Argument embeddings are learned from surrounding contexts involving the predicate and neighboring arguments, while predicate embeddings are learned from argument contexts. The induced represe...
متن کامل