SemTransCNC 1.0 is a semantic transparency dataset of Chinese nominal 
compound which was built using a series of Mechanical Turk-based 
experiments. It consists of the overall and the constituent semantic 
transparency (OST, CST respectively) data of 1,176 dimorphemic Chinese 
nominal compounds which consist of free morphemes and which have mid-range 
frequencies. The dataset is in CSV format which has 11 columns. The 
columns WORD and WORDT list Chinese nominal compounds in simplified and 
traditional Chinese characters respectively. The column STRUCT stores the 
morphological structure of the compounds. The columns FREQ and RFREQ show 
the absolute and relative frequencies respectively of the compounds 
according to the Sinica Corpus V4.0. The columns NOST, NC1CST, and NC2CST 
respectively store the overall semantic transparency value of each 
compound and the constituent semantic transparency values of both 
constituents of each compound; in these columns, 1 means completely 
transparent and 0 means completely opaque. The columns OSD, C1SD, and C2SD 
are the standard deviations of the overall and constituent semantic 
transparency rating data of each compound. Wang et. al. (2014, 2015, 2019) 
describes the details of the construction of the dataset. 

References:

[1] Wang, S.; Huang, C.-R.; Yao, Y. & Chan, A. (2019), The effect of 
morphological structure on semantic transparency ratings, Language and 
Linguistics 20(2), 225--255.
[2] Wang, S.; Huang, C.-R.; Yao, Y. & Chan, A. (2015), Mechanical 
Turk-based experiment vs laboratory-based experiment: A case study on the 
comparison of semantic transparency rating data, in 'Proceedings of the 
29th Pacific Asia Conference on Language, Information and Computation 
(PACLIC-29)', pp. 53--62.
[3] Wang, S.; Huang, C.-R.; Yao, Y. & Chan, A. (2014), Building a semantic 
transparency dataset of Chinese nominal compounds: A practice of 
crowdsourcing methodology, in 'Proceedings of Workshop on Lexical and 
Grammatical Resources for Language Processing (LG-LP 2014 at 
COLING-2014)', Association for Computational Linguistics and Dublin City 
University, Dublin, Ireland, pp. 147--156.