Abstract: This paper attempts to investigate the clustering effect of the emerging intensifiers "chāo (超)", "jù (巨)", "kuáng (狂)", "bào (暴)", "bào (爆)", "qí (奇)" and "zhēnxīn (真心)" through hierarchical cluster analysis. The findings are presented in dendrogram. It is found that these seven intensifiers can be divided into two major categories according to their similarity: "qí" forms its own category, while the other six form a separate one, which can be divided into two subcategories: "chāo" and "zhēnxīn" as one, while "bào (暴)", "bào (爆)", "jù", and "kuáng" as the other. The latter subcategory can be further divided into two groups, "bào (暴)" and "bào (爆)" into one, while "jù" and "kuáng" into the other. It can be seen from the serpentine curve that the disyllabic modified components behind "chāo" and "zhēnxīn" are more likely used to express positive semantic prosody, while "jù" and "kuáng" are more likely to express negative and neutral semantic prosody. The modified components behind "bào (暴)" and "bào (爆)" are more monosyllabic and mostly static. In addition, compared with other adjectives, the adjectives "měng (猛)", "dà (大)", "kuài (快)", "hǎo (好)", "cháng (长)" , "qiáng (强)" , and "duō (多)" have stronger co-occurrence with emerging intensifiers.
Keywords: Intensifiers, hierarchical cluster analysis, semantic prosody, dendrogram, serpentine curve