Context is the only means to identify the sense of a polysemous word. All algorithms for word sense disambiguation make use of information within a context window of the target word. What is the best window size for word sense disambiguation has been long a problem. Different contexts generally give different results even for a same algorithm. In this paper, we exploit an algorithm which is more robust with the varying of different context used. This method aims to lower the uncertainty brought by classifiers using different context window sizes and make more robust utilization of context while perform well. Experiments show our approach outperforms some other algorithms on both robustness and performance.
展开▼
机译:上下文是识别多态词感的唯一意义。 Word Sense消歧的所有算法在目标字的上下文窗口中使用信息。 Word Sense Dismigation的最佳窗口大小是较长的问题。即使对于相同的算法,不同的上下文通常也给出不同的结果。在本文中,我们利用了一种算法,其算法更加强大,随着所使用的不同上下文而变化。此方法旨在使用不同上下文窗口大小降低分类器所带来的不确定性,并在执行良好的情况下进行更强大的情况使用上下文。实验表明我们的方法占鲁棒性和性能的其他一些算法。
展开▼