题名: uyghur semantic similarity computation based on contextual information in web documents
作者: Ma Bo; Zhou Xi; Yang Yating; Zhou Junlin
刊名: Journal of Computational Information Systems
卷: 8, 2, 563-570
英文摘要: In many research fields such as Linguistics, Natural Language Processing, and Artificial Intelligence, semantic similarity computation between words is an important issue. In this paper, semantic similarity metrics are firstly introduced and analyzed in order to determine their advantages and limitations, then a new unsupervised Uyghur context-based semantic similarity metric is proposed combining the feature characteristics of Uyghur. The proposed metric is automatic, do not require any annotated knowledge resources, and can be applied to other languages. In this work, the proposed metric is evaluated on the Miller & Charles data set, the metric is evaluated for different feature weighting schemes and as a function of the number of Web documents used. 50 Uyghur speakers are chosen to take the experiments, it is shown that the correlation scores between context-based similarity metric and human judgments are significantly higher than that of the co-occurrence-based metrics. 1553-9105/Copyright © 2012 Binary Information Press.
Appears in Collections: 多语种信息技术研究室_期刊论文
There are no files associated with this item.
作者单位: Research Center for Multilingual Information Technology, Xinjiang Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Urumuqi 830011, China;Xinjiang Branch of Chinese Academy of Sciences, Urumuqi 830011, China
Ma Bo,Zhou Xi,Yang Yating,et al. uyghur semantic similarity computation based on contextual information in web documents[J]. Journal of Computational Information Systems,2012,8(2):563-570.