Dataset Information

Characterizing the Google Books Corpus: Strong Limits to Inferences of Socio-Cultural and Linguistic Evolution.

ABSTRACT: It is tempting to treat frequency trends from the Google Books data sets as indicators of the "true" popularity of various words and phrases. Doing so allows us to draw quantitatively strong conclusions about the evolution of cultural perception of a given topic, such as time or gender. However, the Google Books corpus suffers from a number of limitations which make it an obscure mask of cultural popularity. A primary issue is that the corpus is in effect a library, containing one of each book. A single, prolific author is thereby able to noticeably insert new phrases into the Google Books lexicon, whether the author is widely read or not. With this understood, the Google Books corpus remains an important data set to be considered more lexicon-like than text-like. Here, we show that a distinct problematic feature arises from the inclusion of scientific texts, which have become an increasingly substantive portion of the corpus throughout the 1900 s. The result is a surge of phrases typical to academic articles but less common in general, such as references to time in the form of citations. We use information theoretic methods to highlight these dynamics by examining and comparing major contributions via a divergence measure of English data sets between decades in the period 1800-2000. We find that only the English Fiction data set from the second version of the corpus is not heavily affected by professional texts. Overall, our findings call into question the vast majority of existing claims drawn from the Google Books corpus, and point to the need to fully characterize the dynamics of the corpus before using these data sets to draw broad conclusions about cultural and linguistic evolution.

SUBMITTER: Pechenick EA

PROVIDER: S-EPMC4596490 | biostudies-literature | 2015

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Characterizing the Google Books Corpus: Strong Limits to Inferences of Socio-Cultural and Linguistic Evolution.

Pechenick Eitan Adam EA Danforth Christopher M CM Dodds Peter Sheridan PS

PloS one 20151007 10

It is tempting to treat frequency trends from the Google Books data sets as indicators of the "true" popularity of various words and phrases. Doing so allows us to draw quantitatively strong conclusions about the evolution of cultural perception of a given topic, such as time or gender. However, the Google Books corpus suffers from a number of limitations which make it an obscure mask of cultural popularity. A primary issue is that the corpus is in effect a library, containing one of each book. ...[more]

PMID: 26445406

Similar Datasets

Project description:From the foods we eat and the houses we construct, to our religious practices and political organization, to who we can marry and the types of games we teach our children, the diversity of cultural practices in the world is astounding. Yet, our ability to visualize and understand this diversity is limited by the ways it has been documented and shared: on a culture-by-culture basis, in locally-told stories or difficult-to-access repositories. In this paper we introduce D-PLACE, the Database of Places, Language, Culture, and Environment. This expandable and open-access database (accessible at https://d-place.org) brings together a dispersed corpus of information on the geography, language, culture, and environment of over 1400 human societies. We aim to enable researchers to investigate the extent to which patterns in cultural diversity are shaped by different forces, including shared history, demographics, migration/diffusion, cultural innovations, and environmental and ecological conditions. We detail how D-PLACE helps to overcome four common barriers to understanding these forces: i) location of relevant cultural data, (ii) linking data from distinct sources using diverse ethnonyms, (iii) variable time and place foci for data, and (iv) spatial and historical dependencies among cultural groups that present challenges for analysis. D-PLACE facilitates the visualisation of relationships among cultural groups and between people and their environments, with results downloadable as tables, on a map, or on a linguistic tree. We also describe how D-PLACE can be used for exploratory, predictive, and evolutionary analyses of cultural diversity by a range of users, from members of the worldwide public interested in contrasting their own cultural practices with those of other societies, to researchers using large-scale computational phylogenetic analyses to study cultural evolution. In summary, we hope that D-PLACE will enable new lines of investigation into the major drivers of cultural change and global patterns of cultural diversity.

Project description:Exchanging gazes with a social partner in response to an event in the environment is considered an effective means to direct attention, share affective experiences, and highlight a target in the environment. This behavior appears during infancy and plays an important role in children's learning and in shaping their socio-emotional development. It has been suggested that cultural values of the community affect socio-emotional development through attentional dynamics of social reference (Rogoff et al., 1993). Maturational processes of brain-circuits have been found to mediate socio-cultural learning and the behavioral manifestation of cultural norms starting at preschool age (Nelson and Guyer, 2011). The aim of the current study was to investigate the relations between cultural ecology levels and children's joint attention (JA). Initiation of JA bids was studied empirically as a function of the level of social load of the target toy (3 levels), the community level of adherence to traditional values (3 levels), parental education (2 levels), and gender. Sixty-two kindergarten aged children were enrolled in a structured toy-exploration task, during which they were presented with toys of various social loads, with social agents (i.e., mother and experimenter) present nearby, and non-social distracters presented intermittently. Measurements included the child's number of JA bids and the extent of positive affect. Analysis of variance indicated that the child's initiation of JA toward the social partner was affected by all levels of cultural ecology (i.e., toy's social load, adherence to tradition values, parental education, gender), thus supporting the study's hypotheses. The effects were such that overall, children, particularly girls' JA initiation was augmented in social toys and moderated by the socio-cultural variables. These results suggest that cultural ecology is related to children's JA, thereby scaffolding initiation of social sharing cues between children and adults. JA plays a role in adjusting children's internal representations of their respective ecological environment.

Dataset Information

Characterizing the Google Books Corpus: Strong Limits to Inferences of Socio-Cultural and Linguistic Evolution.

Publications

Characterizing the Google Books Corpus: Strong Limits to Inferences of Socio-Cultural and Linguistic Evolution.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets