Designing a computer-facilitated learning-tool for learning near-synonyms in Chinese.
Date of Issue2009
School of Electrical and Electronic Engineering
Choosing an appropriate word from a cluster of near-synonyms is a significant challenge in second language acquisition. Second language learners (L2 learners) are found facing difficulties in distinguishing words that have similar meanings. The near-synonyms of physical action verbs (PA Verbs) for instance can be distinctive from each other in the way in which their actions are depicted linguistically. Gao  provided a demonstration of her decomposition method of the semantic properties of sub-classes of near-synonyms of Chinese PA Verbs, where differences among the near-synonyms are identified by investigation of meaning components, marked by different notions found in PA Verbs, such as Bodypart Contact, Instrument, Force, Motion Direction, Speed, Effect, Intention, and Patient Object etc. Based on her study, we propose an e-learning tool to facilitate advanced L2 learners in choosing the most appropriate PA Verb from a group of near-synonyms according to his/her intended context. In our design, we firstly transferred Gao’s semantic decomposition of semantic properties into mathematical expressions, applying two methods: an algorithmic method and a stochastic method. A near-synonyms database was built up based on the quantifications of the semantic properties. Interfaces were created to guide L2 learners in acquiring the most appropriate member from a class of near-synonyms. While the algorithmic method works well in directing learners to a correct synonym in the shortest path, the stochastic method, built with a corpus approach, guarantees a higher certainty in the near-synonym discrimination task. Being first-of-its-kind e-learning tool targeting on assisting L2 learners in near-synonyms acquisition, it provides a good starting point for us to investigate near-synonym acquisition and e-learning tool development.
Final Year Project (FYP)
Nanyang Technological University