基于知网的无指导词义消歧  

An Unsupervised Approach To Word Sense Disambiguation Based on Hownet

作　　者：陈浩[1] CHEN Hao（Department of Computer Science, Guangdong University of Finace and Economics＆, Huashang College, Guangzhou 510000, China）

机构地区：[1]广东财经大学华商学院信息工程系,广东广州510000

出　　处：《电脑知识与技术》2015年第4期67-68,71,共3页Computer Knowledge and Technology

摘　　要：词义消歧仍然是自然语言处理中一个重大的挑战，在自然语言处理的一开始，词义消歧就被认为是自然语言处理的中心任务之一。这篇文章提出了一种无导词义消歧的方法，该方法采用二阶context构造上下文向量，使用k-means算法进行聚类，最后通过计算相似度来进行词义的排歧．实验是在抽取术语的基础上进行的，在多个汉语高频多义词的两组测试中取得了平均准确率82．67％和84．55％的较好的效果。Word sense disambiguation is still as a considered one of the most challenging problems in natural language process- ing. Ever since the field＇ s inception WSD has been perceived as one of the central problems in NLP. This paper presents an unsu- pervised approach which constructs context vector by means of second-order context, clustering by k-means and disambiguates by calculating the similarity. Our experiments are based on the extraction of term and average accuracy is 82.62% and 84.55% for 7 ambiguous words in open test by this method.

关 键 词：词义消歧 HOWNET 二阶context K-MEANS聚类 

分 类 号：TP391[自动化与计算机技术—计算机应用技术;自动化与计算机技术—计算机科学与技术]