Mining the Twentieth Century’s History from the Time Magazine Corpus

Abstract

In this paper we report on an explorative study of the history of the twentieth century from a lexical point of view. As data, we use a diachronic collection of 270,000+ English-language articles harvested from the electronic archive of the well-known Time Magazine (1923Magazine ( -2006. We attempt to automatically identify significant shifts in the vocabulary used in this corpus using efficient, yet unsupervised computational methods, such as Parsimonious Language Models. We offer a qualitative interpretation of the outcome of our experiments in the light of momentous events in the twentieth century, such as the Second World War or the rise of the Internet. This paper follows up on a recent string of frequentist approaches to studying cultural history ('Culturomics'), in which the evolution of human culture is studied from a quantitative perspective, on the basis of lexical statistics extracted from large, textual data sets.

Catching the Red Priest: Using Historical Editions of Encyclopaedia Britannica to Track the Evolution of Reputations

“…Our approach is somewhat similar in that the corpus of historical editions of Britannica can be considered a knowledge base, with an important difference being a chronological dimension, absent from such knowledge bases as YAGO or DBpedia. An example of mining a historical corpus for trends using the frequentist approaches to vocabulary shifts as well as normalization to structured sources can be found in the recent work on newspaper and journal historical editions such as Kestemont et al (2014) and Huet et al (2013).…”

Pandemic news information uncertainty—News dynamics mirror differential response strategies to COVID-19

“…By embedding the co-occurrence structure in a lowdimensional space, it has been shown that newspaper content reflects fundamental cultural movements and understandings from the 19 th century onward [20] and that these contextdependent representations are sensitive to cultural bias as reflected in newspapers [23]. In continuation of the 'Culturomics' movement that used Google Books to show how lexical variation is sensitive to events [24], a wide range of studies has demonstrated that simple word and concept frequencies are sufficient for robust offline detection of major historical events [25] and can be used to model the evolution of complicated cultural processes such as the historical interdependencies between media and politics [26]. Fluctuations of time-dependent word frequencies have been shown to discriminate between classes of events that have class-specific fractal signatures, where the social-cultural class displays non-stationary and on-off intermittent behavior [27].…”

Combining Visual and Textual Features for Semantic Segmentation of Historical Newspapers

“…1 This momentum is particularly vivid in the domain of digitized newspaper archives for which there has been a notable increase of research initiatives over the last years. Those range from individual works dedicated to the development of tools [Yang et al, 2011, Dinarelli and Rosset, 2012, Moreux, 2016, Wevers, 2019 or the usage of those tools [Kestemont et al, 2014, Lansdall-Welfare et al, 2017, to evaluation campaigns [Rigaud et al, 2019, Clausner et al, 2019, including the emergence of large consortia projects seeking to apply computational methods to historical newspapers at scale, such as ViralTexts 2 , Oceanic Exchanges 3 , impresso 4 , NewsEye 5 , and Living with Machines 6 [Ridge et al, 2019].…”