The effect of morphological structure on semantic transparency ratings

Abstract Semantic transparency deals with the interface between lexical semantics and morphology. It is an important linguistic phenomenon in Chinese context of prediction meanings compounds from their constituents. Given prominence compounding morpho-lexical processes, to date there no semantic dataset available support verifiable replicable quantitative analysis Mandarin Chinese. In addition, relation morphological structure has not been systematically examined. This paper reports a crowdsourcing-based experiment designed for construction large which includes ratings both compound each constituent root compound. We also present effects on using constructed dataset. Our study found that transparent modifier-head compound, head tends get greater rating than modifier. Interestingly, such effect observed coordinative compounds. result suggests different structures are processed differently concept plays role word-formation process compounding. advocate crowdsourcing can be highly instrumental method collect judgments construct language resources studies. proposed methodology comparing word sheds light cognitive processing meanings.