The use of language is subject to variation over time as well as across social groups and knowledge domains, leading to differences even in themonolingual scenario . Such variation in word usage is often called lexicalsemantic change (LSC) The goal of LSC is to characterize and quantify languagevariations with respect to word meaning, to measure how distinct two languagesources are . To that end, we propose a self-supervised approach to model LSC by generating training samples by introducing perturbations ofword vectors in the input corpora . We show that our method can be used for thedetection of semantic change with any alignment method . We illustrate the utility of our techniques using experimental results onthree different datasets, involving words with the same or different meanings . We also show that the methods not only provide significant improvements but also can lead tonovel findings for the LSC problem. Our method can also be used to choose the landmark words to use in

Author(s) : Maurício Gruppi, Sibel Adalı, Pin-Yu Chen

Links : PDF - Abstract

Code :

Keywords : lsc - word - method - words - change -

Leave a Reply

Your email address will not be published. Required fields are marked *