Interactive Text Visualization with Text Variation Explorer

Siirtola, Harri; Isokoski, Poika; Säily, Tanja; Nevalainen, Terttu

doi:10.1109/iv.2016.57

“…Therefore, dimension reduction techniques, such as Principal Component Analysis (PCA) or Multidimensional Scaling (MDS) [10], can be used in order to transform a matrix into a two-dimensional scatter plot, a topological landscape or a knowledge map. An example is the Text Variation Explorer [11], which uses PCA to represent various sociolinguistic features in text fragments using a scatter plot. Another example is Memory Islands [12], which integrate hierarchical knowledge, e.g.…”

Section: Discussionmentioning

confidence: 99%

Visualizing Symmetric Square Matrices with Rainbow Boxes: Methods and Application to Character Co-occurrence Matrices in Literary Texts

Lamy

¹

2018

2018 22nd International Conference Information Visualisation (IV)

View full text Add to dashboard Cite

show abstract

“…In future research, the exploratory approach taken in this paper could be further enhanced by interactive visualisation: linking the texts, metadata, visualisations and statistical analyses to each other would greatly facilitate work in historical sociolinguistics. We are already working on this in a project led by Terttu Nevalainen: a second version of our Text Variation Explorer tool (Siirtola et al 2014(Siirtola et al , 2016 will come out in 2017, and another tool for historical sociolinguistic research called Khepri (Mäkelä et al 2016) will be available by the end of 2018.…”

Section: Discussionmentioning

confidence: 99%

Exploring part-of-speech frequencies in a sociohistorical corpus of English

Säily¹,

Vartiainen²,

Siirtola

³

2017

Advances in Historical Sociolinguistics

Self Cite

View full text Add to dashboard Cite

We investigate the usefulness of part-of-speech (POS) annotation as a tool in the study of sociolinguistic variation and genre evolution. We analyse how POS ratios change over time in the Parsed Corpus of Early English Correspondence (c.1410-1681), which social groups lead the changes, and whether the changes can be connected to colloquialisation with regard to reduced complexity or an increasingly involved style. While we find gentryled colloquialisation in terms of noun and verb frequencies as well as evidence for gendered styles, the results on structural complexity are more mixed. We argue that POS annotation can be a useful tool when complemented by a thorough textual analysis, but that more fine-grained categories are needed to reach firmer conclusions.

show abstract

“…In future research, the exploratory approach taken in this paper could be further enhanced by interactive visualisation: linking the texts, metadata, visualisations and statistical analyses to each other would greatly facilitate work in historical sociolinguistics. We are already working on this in a project led by Terttu Nevalainen: a second version of our Text Variation Explorer tool (Siirtola et al 2014(Siirtola et al , 2016 will come out in 2017, and another tool for historical sociolinguistic research called Khepri (Mäkelä et al 2016) will be available by the end of 2018.…”

Section: Discussionmentioning

confidence: 99%

Exploring Future Paths for Historical Sociolinguistics

Säily¹,

Nurmi

²

,

Palander-Collin

³

et al. 2017

Advances in Historical Sociolinguistics

View full text Add to dashboard Cite

In this paper we investigate the usefulness of part-of-speech (POS) annotation in the study of sociolinguistic variation and genre evolution. Our data consist of private letters extracted from the Parsed Corpus of Early English Correspondence, and the material studied covers the period from c.1410 to 1681. By focusing the research on changes in the frequencies of POS labels, we firstly aim to explore the extent to which POS ratios can be used as a tool in diachronic linguistic research (see e.g. Mair et al. 2002). Our second aim is to see if, and how, the genre of personal correspondence changed in the period studied. We are particularly interested in finding out whether the genre has become more informal or "oral" in the course of history by becoming structurally less complex, and thus increasingly corresponding to the norms of spoken language in terms of POS distribution (see e.g. Biber and Finegan 1989, 1992, 1997). Our third topic of inquiry focuses on sociolinguistic variation and sociolinguistically conditioned change. We are interested in seeing whether different discourse styles can be identified in the data by observing differences in POS ratios in texts written by different social groups. In particular, we will focus our attention on the nuclear family in order to test our hypothesis that intimacy between the sender and the recipient of the letter may have facilitated a colloquial style of writing and colloquialisation. In sum, we argue that POS annotation can be a useful tool in the sociohistorical study of genres and genre evolution, but it needs to be complemented by a thorough textual analysis: while we find that the changes in POS ratios correctly predict a gradual colloquialisation of the genre in terms of noun and verb frequencies, for example, they also give some conflicting information about colloquialisation as measured by structural complexity.

show abstract

Interactive Text Visualization with Text Variation Explorer

Cited by 8 publications

References 12 publications

Visualizing Symmetric Square Matrices with Rainbow Boxes: Methods and Application to Character Co-occurrence Matrices in Literary Texts

Visualizing Symmetric Square Matrices with Rainbow Boxes: Methods and Application to Character Co-occurrence Matrices in Literary Texts

Exploring part-of-speech frequencies in a sociohistorical corpus of English

Exploring Future Paths for Historical Sociolinguistics

Contact Info

Product

Resources

About