Linguistics Colloquium: Eitan Grossman
Haim Dubossarsky (HUJI ELSC) & Eitan Grossman (HUJI Linguistics & LLCC)
Title: Computational accounts of semantic change
Abstract: Linguists have identified a number of types of recurrent semantic change, and have proposed a number of explanations, usually based on specific lexical items. Our research takes a different approach, by using word embedding and a distributional semantic model to identify and quantify semantic change over time across an entire lexicon in a completely bottom-up fashion, and by examining which properties of words are causal factors in semantic change. Altogether, the results indicate that semantic change varies considerably through time. Although words change stochastically, we found there are marked periods in which words exhibit different patterns of distributional properties, in addition to periods of increased or decreased semantic change. Moreover, several independent contributing factors for semantic change in words are identified. First, the degree of prototypicality of a word within its semantic cluster correlates inversely with its likelihood of change (the “Diachronic Prototypicality Effect”). Second, the word class assignment of a word correlates with its rate of change: verbs change more than nouns, and nouns change more than adjectives (the “Diachronic Word Class Effect”), which we propose may be the diachronic result of an independently established synchronic psycholinguistic effect (the “Verb Mutability Effect”).
Building 604, room 11