An Antonym Study Of Irony Calculations

With the development of language information processing and knowledge engineering, the importance and urgency of automatic semantic analysis become increasingly apparent. Irony is a common figures of speech whose literal meaning is opposite to it’s implied meaning, irony is often applied in the field of both written and oral language. At present, the study of irony processing focus to automatically detecting verbal irony, the difficulty of irony automatic understanding and generation is that though the computer know irony statements are used to express the contrary of what is being said, meanwhile, the words have different antonyms in different contexts, so that the actual meaning of the sentence expression of irony is vague. Therefore, we need available resources to solve this problem, and to provide the basic resources to irony automatically research.A word in different contexts have different antonyms, such as tall/low, tall/short. The context of antonyms related with their modified terms to a great extent. Some scholars have pointed out that we can study antonyms by observing the characteristics of the nouns which modified by antonyms adjective. This characteristic can be summarized as adjective words’selective restriction to the modified words. Accordingly, this article attempts to observe the selective restriction of antonym adjectives, examining whether the selective restriction as a condition of opposite relationship, and exploring how to choose the most appropriate antonyms in different contexts. Launching basic research for getting cater to irony automatically generating and understanding antonym resources.Chinese antonyms is in large amount, in order to make the object more typical and representative, this paper selected some high-frequency adjective from Modern Chinese Frequency Dictionary, found out the antonyms of these high-frequency adjective referred to the Antonym Dictionary. And then, in order to inspect the adjective restrictions to nouns, the writer get words that can be modified by antonyms based on the Chinese CogBank, Using HowNet’s sememe system to describe how’s antonymous adjectives apply selective restriction to its’modified words, then to observe the function that semantic class plays to revealing the condition that antonyms relation.By examining the 98 adjectives (93 antonyms) semantic choice restrictions to the words they modified, we can get the conclusion:In the selective restriction, semantic class constraints can be the condition of antonym relationship for some words, that is the opposite of semantic-sensitive. We can use the semantic class of the modified words as the most appropriate basis to choose its antonym in different contexts. The insensitive semantic class antonyms, other strategies need to be taken to examine their choice of the most appropriate antonyms conditions in different contexts, such as, further study of limiting factor of the words it modifies like semantic properties characteristics.