Statistically Significant Detection of Linguistic Change

Abstract

We propose a new computational approach for tracking and detecting statistically significant linguistic shifts in the meaning and usage of words. Such linguistic shifts are especially prevalent on the Internet, where the rapid exchange of ideas can quickly change a word's meaning. Our meta-analysis approach constructs property time series of word usage, and then uses statistically sound change point detection algorithms to identify significant linguistic shifts.We consider and analyze three approaches of increasing complexity to generate such linguistic property time series, the culmination of which uses distributional characteristics inferred from word co-occurrences. Using recently proposed deep neural language models, we first train vector representations of words for each time period. Second, we warp the vector spaces into one unified coordinate system. Finally, we construct a distance-based distributional time series for each word to track it's linguistic displacement over time.We demonstrate that our approach is scalable by tracking linguistic change across years of micro-blogging using Twitter, a decade of product reviews using a corpus of movie reviews from Amazon, and a century of written books using the Google Book-ngrams. Our analysis reveals interesting patterns of language usage change commensurate with each medium.

Exploring Diachronic Lexical Semantics with JeSemE

“…Future technical work will add functionality to compare words across corpora which might require a mapping between embeddings (Kulkarni et al, 2015;Hamilton et al, 2016) and provide optional stemming routines. Both goals come with an increase in precomputed similarity values and will thus necessitate storage optimizations to ensure long-term availability.…”

“…We follow Kim et al (2014) in choosing such a visualization, while we refrain from using the two-dimensional projection used in other studies (Kulkarni et al, 2015;Hamilton et al, 2016). We stipulate that the latter could Figure 2: Screenshot of JESEME's result page when searching for the lexical item "heart" in COHA.…”

Tracing armed conflicts with diachronic word embedding models

“…Unfortunately, there is no such manually compiled compendium of all the semantic shifts that English words underwent in the last two centuries. The problem is even more severe for studies using more fine-grained time units spanning days or years, rather than decades, like in (Kulkarni et al, 2015) or (Kutuzov and Kuzmenko, 2016): When trying to uncover subtle changes of perspective (for example, 'Trump' moving towards being associated with 'president' rather than 'millionaire'), it is difficult to find gold standard annotations for rigorous evaluation of the proposed methods.…”

Adapting predominant and novel sense discovery algorithms for identifying corpus-specific sense differences

“…It is an important as well as challenging task to identify predominant word senses specific to various corpora. While the researchers have started exploring the temporal and spatial scopes of word senses (Cook and Stevenson, 2010;Gulordava and Baroni, 2011;Kulkarni et al, 2015;Jatowt and Duh, 2014;Mitra et al, 2014;Mitra et al, 2015), corpora-specific senses have remained mostly unexplored. Our contributions: Motivated by the above applications, this paper studies corpora-specific senses for the first time and makes the following contributions 1 : (i) we take two different meth- 1 The code and evaluation results are available at: http: //tinyurl.com/h4onyww ods for novel sense discovery (Mitra et al, 2014;Lau et al, 2014) and one for predominant sense identification (McCarthy et al, 2004) and adapt these in an automated and unsupervised manner to identify corpus-specific sense for a given word (noun), and (ii) perform a thorough manual evaluation to rigorously compare the corpus-specific senses obtained using these methods.…”