Gaussian Mixture Embeddings for Multiple Word Prototypes
This work addresses the polysemy problem in natural language processing for tasks requiring accurate word semantics, representing an incremental improvement over existing multiple vector methods.
The paper tackles the problem of polysemy in word representation by proposing Gaussian mixture embeddings, where each word is modeled as a Gaussian mixture distribution with each component representing a word sense, and experiments on four benchmarks demonstrate its effectiveness.
Recently, word representation has been increasingly focused on for its excellent properties in representing the word semantics. Previous works mainly suffer from the problem of polysemy phenomenon. To address this problem, most of previous models represent words as multiple distributed vectors. However, it cannot reflect the rich relations between words by representing words as points in the embedded space. In this paper, we propose the Gaussian mixture skip-gram (GMSG) model to learn the Gaussian mixture embeddings for words based on skip-gram framework. Each word can be regarded as a gaussian mixture distribution in the embedded space, and each gaussian component represents a word sense. Since the number of senses varies from word to word, we further propose the Dynamic GMSG (D-GMSG) model by adaptively increasing the sense number of words during training. Experiments on four benchmarks show the effectiveness of our proposed model.