A Wind of Change: Detecting and Evaluating Lexical Semantic Change across Times and Domains

@inproceedings{Schlechtweg2019AWO,
  title={A Wind of Change: Detecting and Evaluating Lexical Semantic Change across Times and Domains},
  author={Dominik Schlechtweg and Anna H{\"a}tty and Marco Del Tredici and Sabine Schulte im Walde},
  booktitle={Annual Meeting of the Association for Computational Linguistics},
  year={2019}
}
We perform an interdisciplinary large-scale evaluation for detecting lexical semantic divergences in a diachronic and in a synchronic task: semantic sense changes across time, and semantic sense changes across domains. Our work addresses the superficialness and lack of comparison in assessing models of diachronic lexical change, by bringing together and extending benchmark models on a common state-of-the-art evaluation task. In addition, we demonstrate that the same evaluation task and… 
View on ACL
arxiv.org
74 Citations
Highly Influential Citations
9
Background Citations
45
Methods Citations
36
Results Citations
9

Tables from this paper

74 Citations

Time-Out: Temporal Referencing for Robust Modeling of Lexical Semantic Change

This work empirically tested the Temporal Referencing method for lexical semantic change and showed that, trained on a diachronic corpus, the skip-gram with negative sampling architecture with temporal referencing outperforms alignment models on a synthetic task as well as a manual testset.

Lexical Semantic Change Discovery

By heavily fine-tuning a type-based and a token-based approach on recently published German data, it is demonstrated that both models can be applied to discover new words undergoing meaning change and provided an almost fully automated framework for both evaluation and discovery.

Simulating Lexical Semantic Change from Sense-Annotated Data

This work presents a novel procedure to simulate lexical semantic change from synchronic sense-annotated data, and provides the first large-scale evaluation gold standard for LSC detection models.

SemEval-2020 Task 1: Unsupervised Lexical Semantic Change Detection

This work presents the results of the first shared task that addresses the most pressing problem in Lexical Semantic Change detection, by providing researchers with an evaluation framework and manually annotated, high-quality datasets for English, German, Latin, and Swedish.

Shared Task: Lexical Semantic Change Detection in German

The results of the first shared task on unsupervised lexical semantic change detection (LSCD) in German based on the evaluation framework proposed by Schlechtweg et al. (2019) are presented.

Autoencoding Word Representations through Time for Semantic Change Detection

This work proposes three variants of sequential models for detecting semantically shifted words, effectively accounting for the changes in the word representations over time, in a temporally sensitive manner and takes a step towards comparing different approaches in a quantitative manner.

Effects of Pre- and Post-Processing on type-based Embeddings in Lexical Semantic Change Detection

This work optimize existing models by pre-training on large corpora and refining on diachronic target corpora tackling the notorious small data problem, and applying post-processing transformations that have been shown to improve performance on synchronic tasks.

Sequential Modelling of the Evolution of Word Representations for Semantic Change Detection

This work proposes three variants of sequential models for detecting semantically shifted words, effectively accounting for the changes in the word representations over time, and compares different approaches in a quantitative manner, demonstrating that temporal modelling of word representations yields a clear-cut advantage in performance.

Scalable and Interpretable Semantic Change Detection

This work proposes a novel scalable method for word usage-change detection that offers large gains in processing time and significant memory savings while offering the same interpretability and better performance than unscalable methods.

Analyzing Gaussian distribution of semantic shifts in Lexical Semantic Change Models

This work compares state-of-the-art computational historical linguistics approaches to evaluate the efficacy of thresholds based on the Gaussian Distribution of semantic shifts and presents the results of an in-depth analysis conducted on both SemEval-2020 Task 1 Subtask 1 and DIACR-Ita tasks.
...

References

SHOWING 1-10 OF 73 REFERENCES

Time-Out: Temporal Referencing for Robust Modeling of Lexical Semantic Change

This work empirically tested the Temporal Referencing method for lexical semantic change and showed that, trained on a diachronic corpus, the skip-gram with negative sampling architecture with temporal referencing outperforms alignment models on a synthetic task as well as a manual testset.

A framework for analyzing semantic change of words across time

An exploratory analysis aiming to investigate methods for studying and visualizing changes in word meaning over time, and proposes a framework for exploring semantic change at the lexical level, at the contrastive-pair level, and at the sentiment orientation level.

A Bayesian Model of Diachronic Meaning Change

A dynamic Bayesian model of diachronic meaning change is presented, which infers temporal word representations as a set of senses and their prevalence and reveals that it performs on par with highly optimized task-specific systems.

Automatically Identifying Changes in the Semantic Orientation of Words

A web-based method for determining semantic orientation is adapted to the task of identifying ameliorations and pejorations in corpora from differing time periods, and evidence is found that it is able to identify changes in semantic orientation.

Survey of Computational Approaches to Diachronic Conceptual Change

SURel: A Gold Standard for Incorporating Meaning Shifts into Term Extraction

SURel, a novel dataset with human-annotated meaning shifts between general-language and domain-specific contexts, is introduced and it is demonstrated that meaning shifts of term candidates cause errors in term extraction, and the SURel annotation reflects these errors.

Semantic Variation in Online Communities of Practice

It is shown that while some meaning shifts are shared across related communities, others are community-specific, and therefore independent from the discussed topic, and proposed as evidence in favour of sociolinguistic theories of socially-driven semantic variation.

Understanding semantic change of words over centuries

This paper proposes to identify clusters of topics surrounding the entity over time using Topics-Over-Time (TOT) and k-means clustering and finds that the period identified by the model precisely coincides with events in the same period that correspond to the change that occurs.

German in Flux: Detecting Metaphoric Change via Word Entropy

The information-theoretic measure entropy is explored to detect metaphoric change, transferring ideas from hypernym detection to research on language change, and the first diachronic test set for German is built as a standard for metaphoric changes annotation.

Detecting Domain-Specific Ambiguities: An NLP Approach Based on Wikipedia Crawling and Word Embeddings

A natural language processing (NLP) approach is applied based on Wikipedia crawling and word embeddings, a novel technique to represent the meaning of words through compact numerical vectors that allows an estimate of the variation of meaning of the computer science words when used in different domains.
...