Description
Word embeddings – vector representations of words that embed words in a so-called semantic space where the vectors of semantically similar words lie close together – are increasingly used for semantic searches in large text corpora. Word vector distances can be used to build semantic networks of words. This closely resembles the notion of semantic fields that humanities scholars are familiar with. We have been working on an implementation of word embeddings, as produced by a popular implementation word2vec, to trace concepts through time without the dependency of particular keywords (Kenter et.al. 2014). In this paper, we aim to show how this technique can add to existing traditions in conceptual history by providing data-driven insights into semantic change. At the same time, we will address some important challenges that come with the use of word embeddings to represent concepts and conceptual change for the study of history. The use of computational techniques like word2vec demands choices of practical or technical nature. How do we legitimize these choices in terms of conceptual theory? Another problem relates to the dependency on data. Do the results of word embedding techniques provide insights into real conceptual change, or do they merely reflect arbitrary biases in the underlying data? Both challenges illustrate the need for critical reflection now that advanced computational tools are adopted in historical scholarship. Based on concrete examples, we will show how we dealt with these challenges in our research.| Period | 8 Dec 2017 |
|---|---|
| Event title | Genealogies of Knowlegde I: Translating Political and Scientific Thought across Time and Space |
| Event type | Conference |
| Location | Manchester, United KingdomShow on map |
| Degree of Recognition | International |