An Adaptive Hierarchical Compositional Model for Phrase Embedding

Phrase embedding aims at representing phrases in a vector space and it is important for the performance of many NLP tasks. Existing models only regard a phrase as either full-compositional or non-compositional, while ignoring the hybrid-compositionality that widely exists, especially in long phrases. This drawback prevents them from having a deeper insight into the semantic structure for long phrases and as a consequence, weakens the accuracy of the embeddings. In this paper, we present a novel method for jointly learning compositionality and phrase embedding by adaptively weighting different compositions using an implicit hierarchical structure. Our model has the ability of adaptively adjusting among different compositions without entailing too much model complexity and time cost. To the best of our knowledge, our work is the first effort that considers hybrid-compositionality in phrase embedding. The experimental evaluation demonstrates that our model outperforms state-of-the-art methods in both similarity tasks and analogy tasks.

• Publication year

  2018

• Venue

  International Joint Conference on Artificial Intelligence

• Publication date

  2018-07-01

• Fields of study

  Computer Science

• Identifiers
  DOI 10.24963/ijcai.2018/576
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Exploiting Word Internal Structures for Generic Chinese Sentence Representation

  2017cited by this paper
• Adaptive Joint Learning of Compositional and Non-Compositional Phrase Embeddings

  2016cited by this paper
• Unsupervised Word and Dependency Path Embeddings for Aspect Term Extraction

  2016cited by this paper
• The mechanism of additive composition

  2015cited by this paper
• A Multiword Expression Data Set: Annotating Non-Compositionality and Conventionalization for English Noun Compounds

  2015cited by this paper
• Learning Composition Models for Phrase Embeddings

  2015cited by this paper
• TR9856: A Multi-word Term Relatedness Benchmark

  2015cited by this paper
• Learning Semantic Composition to Detect Non-compositionality of Multiword Expressions

  2015cited by this paper
• Dependency-Based Word Embeddings

  2014cited by this paper
• Resolving Lexical Ambiguity in Tensor Regression Models of Meaning

  2014cited by this paper
• Bilingually-constrained Phrase Embeddings for Machine Translation

  2014cited by this paper
• GloVe: Global Vectors for Word Representation

  2014cited by this paper
• Detecting Compositionality of Multi-Word Expressions using Nearest Neighbours in Vector Space Models

  2013cited by this paper
• Computing term similarity by large probabilistic isA knowledge

  2013cited by this paper
• Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank

  2013cited by this paper
• Exploiting Similarities among Languages for Machine Translation

  2013influential reference
• Parsing with Compositional Vector Grammars

  2013cited by this paper
• Distributed Representations of Words and Phrases and their Compositionality

  2013influential reference
• Semantic Compositionality through Recursive Matrix-Vector Spaces

  2012cited by this paper
• Composition in Distributional Models of Semantics

  2010cited by this paper
• Mathematical Foundations for a Compositional Distributional Model of Meaning

  2010cited by this paper
• Nouns are Vectors, Adjectives are Matrices: Representing Adjective-Noun Constructions in Semantic Space

  2010cited by this paper
• Et al

  2008influential reference
• Combining Symbolic and Distributional Models of Meaning

  2007cited by this paper
• Detecting Compositionality of Verb-Object Combinations using Selectional Preferences

  2007cited by this paper
• Measuring the Relative Compositionality of Verb-Noun (V-N) Collocations by Integrating Features

  2005cited by this paper
• Information Theory, Inference, and Learning Algorithms

  2004cited by this paper
• What is Cognitive Science

  1999influential reference

• Parsing and encoding interactive phrase structure for implicit discourse relation recognition

  2024cites this paper
• Bilingual phrase induction with local hard negative sampling

  2024cites this paper
• Enhancing Idiomatic Representation in Multiple Languages via an Adaptive Contrastive Triplet Loss

  2024cites this paper
• Unified Representation for Non-compositional and Compositional Expressions

  2023cites this paper
• A novel topic clustering algorithm based on graph neural network for question topic diversity

  2023cites this paper
• Text-based emotion recognition using contextual phrase embedding model

  2023cites this paper
• Self-supervised phrase embedding method by fusing internal and external semantic information of phrases

  2022cites this paper
• Getting BART to Ride the Idiomatic Train: Learning to Represent Idiomatic Expressions

  2022cites this paper
• Knowledge Is Flat: A Seq2Seq Generative Framework for Various Knowledge Graph Completion

  2022cites this paper
• MDERank: A Masked Document Embedding Rank Approach for Unsupervised Keyphrase Extraction

  2021influential citation
• Phrase embedding learning from internal and external information based on autoencoder

  2021cites this paper
• Phrase2Vec: Phrase embedding based on parsing

  2020cites this paper
• HAMNER: Headword Amplified Multi-span Distantly Supervised Method for Domain Specific Named Entity Recognition

  2019cites this paper