The good, the bad, and the ambivalent: Extrapolating affective values for 38,000+Chinese words via a computational model

Word affective ratings are important tools in psycholinguistic research, natural language processing, and many other fields. However, even for well-studied languages, such norms are usually limited in scale. To extrapolate affective (i.e., valence and arousal) values for words in the SUBTLEX-CH database (Cai & Brysbaert, 2010, PLoS ONE, 5(6):e10729), we implemented a computational neural network which captured how words' vector-based semantic representations corresponded to the probability densities of their valence and arousal. Based on these probability density functions, we predicted not only a word's affective values, but also their respective degrees of variability that could characterize individual differences in human affective ratings. The resulting estimates of affective values largely converged with human ratings for both valence and arousal, and the estimated degrees of variability also captured important features of the variability in human ratings. We released the extrapolated affective values, together with their corresponding degrees of variability, for over 38,000 Chinese words in the Open Science Framework (https://osf.io/s9zmd/). We also discussed how the view of embodied cognition could be illuminated by this computational model.

Automated summary

Word affective ratings play a crucial role in psycholinguistic research and natural language processing. However, the existing norms for affective ratings are often limited in scale. In this study, a computational neural network was implemented to extrapolate the affective values of Chinese words based on their vector-based semantic representations. The resulting estimates of affective values correlated well with human ratings and also captured the variability in human ratings. The extrapolated affective values for over 38,000 Chinese words were made available in the Open Science Framework.