Abstract
Whether a committed reader or not, it is clear that the English language has evolved throughout the years into many different nuances. What is considered acceptable in terms of written form has been adjusted to match cultural, regional, and on a macroscopic scale, temporal changes. This paper explores these changes through innovative analysis of semantics, lexicology, syntax, and context, (to be referred to as Linguistic Patterns), analyzing data and deriving conclusions. It is apparent that there are noticeable differences in the corpora of a 16th-century author to a contemporary one. The most accessible examples of such changes are spotted in written works, whether poetry, books, or documents, as they provide valuable insight into the analysis of change. The methodology incorporates pre-existing and specifically trained models specialized in data analysis to understand and compile these changes. This study showcases the evolution of the English language interpreted by Machine Learning (ML) models and methods such as Natural Language Understanding (NLU). By feeding data, specifically written works, into such models with the foresight of expecting a wide range of differing results and analyzing the changes through the scope of time, this study showcases the change of Linguistic Patterns. The decision between model preference and proficiency is made by comparing the quality of data outputs, and systematically evaluating different model archetypes, such as Generative Pre-training Transformers (GPT) or Bidirectional Encoder Representations from Transformers (BERT). The evaluation of changes in linguistic patterns is quantifiable through statistical measures, embeddings and syntactic parsing scores. Through these steps, this study derives that the English language has experienced a robust alteration in its core, from the elimination of now-considered archaic lexicology, differences in structural and contextual cues as well as notable evolution in semantics. These findings can be utilized in historical linguistic analysis and education, as well as improving Natural Language Understanding.