Non-Linear Similarity Learning for Compositionality
نویسندگان
چکیده
Many NLP applications rely on the existence of similarity measures over text data. Although word vector space models provide good similarity measures between words, phrasal and sentential similarities derived from composition of individual words remain as a difficult problem. In this paper, we propose a new method of of non-linear similarity learning for semantic compositionality. In this method, word representations are learned through the similarity learning of sentences in a high-dimensional space with kernel functions. On the task of predicting the semantic relatedness of two sentences (SemEval 2014, Task 1), our method outperforms linear baselines, feature engineering approaches, recursive neural networks, and achieve competitive results with long short-term memory models.
منابع مشابه
Combining Different Features of Idiomaticity for the Automatic Classification of Noun+Verb Expressions in Basque
We present an experimental study of how different features help measuring the idiomaticity of noun+verb (NV) expressions in Basque. After testing several techniques for quantifying the four basic properties of multiword expressions or MWEs (institutionalization, semantic non-compositionality, morphosyntactic fixedness and lexical fixedness), we test different combinations of them for classifica...
متن کاملLearning Semantic Composition to Detect Non-compositionality of Multiword Expressions
Non-compositionality of multiword expressions is an intriguing problem that can be the source of error in a variety of NLP tasks such as language generation, machine translation and word sense disambiguation. We present methods of non-compositionality detection for English noun compounds using the unsupervised learning of a semantic composition function. Compounds which are not well modeled by ...
متن کاملAdaptive Joint Learning of Compositional and Non-Compositional Phrase Embeddings
We present a novel method for jointly learning compositional and noncompositional phrase embeddings by adaptively weighting both types of embeddings using a compositionality scoring function. The scoring function is used to quantify the level of compositionality of each phrase, and the parameters of the function are jointly optimized with the objective for learning phrase embeddings. In experim...
متن کاملDetecting Compositionality of English Verb-Particle Constructions using Semantic Similarity
We present a novel method for detecting the compositionality of English verbparticle constructions (VPCs), based on the assumption that compositionality can be modelled with semantic similarity between VPCs and their base verb in isolation. We also evaluate the contribution of components in compositional VPCs using semantic overlap.
متن کاملDAG-Structured Long Short-Term Memory for Semantic Compositionality
Recurrent neural networks, particularly long short-term memory (LSTM), have recently shown to be very effective in a wide range of sequence modeling problems, core to which is effective learning of distributed representation for subsequences as well as the sequences they form. An assumption in almost all the previous models, however, posits that the learned representation (e.g., a distributed r...
متن کامل