Temporal Analysis of Language through Neural Language Models

TL;DR

This paper introduces a neural language model trained on historical text data to automatically detect and analyze language change over time, pinpointing when specific words have evolved in meaning from 1900 to 2009.

Contribution

It presents a novel method for temporal language analysis using chronologically trained neural models on large corpora, enabling detection of word meaning changes and their timing.

Findings

Identified significant language changes in words like 'cell' and 'gay' from 1900 to 2009.

Successfully pinpointed specific years when words underwent semantic shifts.

Demonstrated the effectiveness of neural models in temporal linguistic analysis.

Abstract

We provide a method for automatically detecting change in language across time through a chronologically trained neural language model. We train the model on the Google Books Ngram corpus to obtain word vector representations specific to each year, and identify words that have changed significantly from 1900 to 2009. The model identifies words such as "cell" and "gay" as having changed during that time period. The model simultaneously identifies the specific years during which such words underwent change.