Measuring semantic similarity in WordNet

Semantic similarity between words is a generic problem for many applications of computational linguistics and artificial intelligence. The difficulty of this task lies in how to find an effective way to simulate the process of human judgment of word similarity by combining and processing a number of information sources. This paper presents a novel model to measure semantic similarity between words in the WordNet, using edge-counting techniques. The fundamental idea of this model is based on the assumption that human judgment process for semantic similarity can be simulated by the ratio of common features to the total features between words. According to the experiment against a benchmark set by human similarity judgment, our measure achieves a better result. The correlation is 0.926 with average human judgment on a standard 28 word-pair dataset, which outperforms other previous reported methods.