Gating Mechanisms for Combining Character and Word-level Word Representations: an Empirical Study
Explore this paper's citation graph
Summary
There is strong empirical evidence that modeling characters improves the learned representations at the word and sentence levels, and that doing so is particularly useful when representing less frequent words.
- Type
- preprint
- Published
- 2019-04-11
- Cited by
- 3
- References
- 83
- Access
- Open access
- OpenAlex
- https://openalex.org/W2938741373
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:119117163
Keywords
Word (group theory), Computer science, Sentence, Natural language processing, Character (mathematics)
References
- Python: the tutorial
- A large annotated corpus for learning natural language inference
- SimLex-999: Evaluating Semantic Models With (Genuine) Similarity Estimation
- Finding Function in Form: Compositional Character Models for Open Vocabulary Word Representation
- Character-Aware Neural Language Models
- Matplotlib: A 2D Graphics Environment
- Annotating Expressions of Opinions and Emotions in Language
- A word at a time: computing word relatedness using temporal semantic analysis
- The University of South Florida free association, rhyme, and word fragment norms
- A large-scale classification of English verbs
- Placing search in context: the concept revisited
- Long Short-Term Memory
- Learning Question Classifiers
- Contextual correlates of synonymy
- Multimodal Distributional Semantics
- A Sentimental Education: Sentiment Analysis Using Subjectivity Summarization Based on Minimum Cuts
- A unified architecture for natural language processing: deep neural networks with multitask learning
- Collecting Image Annotations Using Amazon’s Mechanical Turk
- Large-scale learning of word relatedness with constraints
- Speech recognition with deep recurrent neural networks
Cited by
Related papers
- A CHARACTER IS A CHARACTER IS A CHARACTER
- A Small Character, A Great Spirit
- Mining user similarity from semantic trajectories
- Computing method for semantic similarity of words based on How Net
- A Hybrid Approach for Measuring Semantic Similarity between Documents and its Application in Mining the Knowledge Repositories
- The Research of Word Similarity in Semantic Retrieval