How to visualize high-dimensional data: a roadmap

Visualization and Data Analysis 2006, 2006

The current availability of information many times impair the tasks of searching, browsing and analyzing information pertinent to a topic of interest. This paper presents a methodology to create a meaningful graphical representation of documents corpora targeted at supporting exploration of correlated documents. The purpose of such an approach is to produce a map from a document body on a research topic or field based on the analysis of their contents, and similarities amongst articles. The document map is generated, after text pre-processing, by projecting the data in two dimensions using Latent Semantic Indexing. The projection is followed by hierarchical clustering to support sub-area identification. The map can be interactively explored, helping to narrow down the search for relevant articles. Tests were performed using a collection of documents pre-classified into three research subject classes: Case-Based Reasoning, Information Retrieval, and Inductive Logic Programming. The map produced was capable of separating the main areas and approaching documents by their similarity, revealing possible topics, and identifying boundaries between them. The tool can deal with the exploration of inter-topics and intra-topic relationship and is useful in many contexts that need deciding on relevant articles to read, such as scientific research, education, and training.

The current availability of information many times impair the tasks of searching, browsing and analyzing information pertinent to a topic of interest. This paper presents a methodology to create a meaningful graphical representation of documents corpora targeted at supporting exploration of correlated documents. The purpose of such an approach is to produce a map from a document body on a research topic or field based on the analysis of their contents, and similarities amongst articles. The document map is generated, after text pre-processing, by projecting the data in two dimensions using Latent Semantic Indexing. The projection is followed by hierarchical clustering to support sub-area identification. The map can be interactively explored, helping to narrow down the search for relevant articles. Tests were performed using a collection of documents pre-classified into three research subject classes: Case-Based Reasoning, Information Retrieval, and Inductive Logic Programming. The map produced was capable of separating the main areas and approaching documents by their similarity, revealing possible topics, and identifying boundaries between them. The tool can deal with the exploration of inter-topics and intra-topic relationship and is useful in many contexts that need deciding on relevant articles to read, such as scientific research, education, and training.

Journal of Data Mining & Digital Humanities, 2018

This article explores whether and how network visualization can benefit philological and historical-linguistic study. This is illustrated with a corpus-based investigation of scribes' language use in a lemmatized and morphologically annotated corpus of documentary Latin (Late Latin Charter Treebank, LLCT2). We extract four continuous linguistic variables from LLCT2 and utilize a gradient colour palette in Gephi to visualize the variable values as node attributes in a trimodal network which consists of the documents, writers, and writing locations underlying the same corpus. We call this network the "LLCT2 network". The geographical coordinates of the location nodes form an approximate map, which allows for drawing geographical conclusions. The linguistic variables are examined both separately and as a sum variable, and the visualizations presented as static images and as interactive Sigma.js visualizations. The variables represent different domains of language competen...

2013

Information visualization is breaking new ground within the fields of the humanities and social sciences, bringing in graphical representations and visual analysis methods for diverse data and topics, which have been verbalized mainly before. Yet, various visualization methods are doing so in a rather basic and selective style, not always tackling their subject matters on an appropriate level of conceptual complexity, leaving their visual results widely unconnected and dynamically underdeveloped. To tackle some of these issues, this article is taking on the information visualization framework of time geography, to focus on the elaboration of its visual elements and their interconnectivity with complementing visualization domains. By outlining possible couplings to methods of process visualization, as well as dynamic social and semantic network visualization, a visual knowledge representation framework is sketched out, whose potentials for illustrating diverse topics of humanities and social science education and research will be discussed.

An advantage of corpus-based language studies is that global descriptions of linguistic texts can be obtained by examining a broad range of linguistic features. However, multivariate statistical techniques are required to analyze the multiple linguistic features found in a number of texts. This study compared the strengths and weaknesses of several multivariate statistical techniques, thereby demonstrating the effectiveness of using heat map with hierarchical clustering as a powerful method for visualizing multivariate data. Explanations are also provided for how these techniques can be used in the R programming language as well as indicating how the results obtained can be interpreted.* 1. Introduction An advantage of corpus-based language studies is that global descriptions of linguistic texts can be obtained by examining a broad range of linguistic features. However, multivariate statistical techniques are required to analyze the multiple linguistic features found in a number of texts. These techniques can identify the complex interrelationships among linguistic features and texts, as well as the associated patterns between linguistic features and texts. The appropriate-ness of these techniques for analyzing language has been confirmed in the past few decades. For example, Burrows (1987) employed principal components analysis to analyze the idiolects of the major characters in Jane Austen's novels. Biber (1988) applied factor analysis to describe the linguistic characteristics of speech and writing. Nakamura and Sinclair (1995) used Hayashi's Quantification Method Type III, a method that is mathematically identical to correspondence analysis, to compare the frequencies of the collocates of the word woman in four components of the Bank of English, a representative subset of the COBUILD Corpus. Moreover, Hoover (2003) used cluster analysis to distinguish texts written by different authors using commonly occurring words in these texts. In the present study, I compared the strengths and weaknesses of several multivariate statistical techniques, thereby demonstrating the effectiveness of using heat map with hierarchical clustering as a powerful method for visualizing multivariate data. I also explain how these techniques are used with R (Ihaka and Gentleman 1996) and how the results can be interpreted.

2000

This article describes the Compus visualization system that assists in the exploration and analysis of structured document corpora encoded in XML. Compus has been developed for and applied to a corpus of 100 French manuscript letters of the 16 th century, transcribed and encoded for scholarly analysis using the recommendations of the Text Encoding Initiative. By providing a synoptic visualization of a corpus and allowing for dynamic queries and structural transformations, Compus assists researchers in finding regularities or discrepancies, leading to a higher level analysis of historic source. Compus can be used with other richly encoded text corpora as well.

Case Studies of …, 1998

In: N. Tahmasebi, L. Borin, A. Jatowt & alii (eds.), Proceedings of the 1st International Workshop on Computational Approaches to Historical Langauge Change. Stroudsbourg: Association for Computational Linguistics (ACL), 109–114., 2019

The study of language change through parallel corpora can be advantageous for the analysis of complex interactions between time, text domain and language. Often, those advantages cannot be fully exploited due to the sparse but high-dimensional nature of such historical data. To tackle this challenge, we introduce ParHistVis: a novel, free, easy-to-use, interactive visualization tool for parallel, multilingual , diachronic and synchronic linguistic data. We illustrate the suitability of the components of the tool based on a use case of word order change in Romance wh-interrogatives.

Proceedings of VisLR, Visualization as Added Value in the Development, Use and Evaluation of Language Resources; Workshop at the 9th Languages Resources Conference (LREC 2014), Reykjavík, 23-26 May 2014 (Workshop Proceedings ed. A. Hautli-Janisz, V. Lyding & C. Rohrdantz), pp 33-40., 2014

We present an innovative visualization technique for the analysis of historical data. We illustrate our method with respect to a diachronic case study involving V1 word order in Icelandic. A number of interacting factors have been proposed by linguists as being determinative of matrix declarative V1. The significance of these factors in contributing to declarative V1 can be explored interactively via our multifactorial visualization within a given text, but also comparatively over time. We apply the visualization to a corpus study based on the IcePaHC historical corpus of Icelandic and show that new results emerge very clearly out of the visualization component and that the appearance of declarative V1 is not confined to the situations identified so far by linguists. We demonstrate that the multifactorial visualization opens up new avenues for the exploration of alternative explanations. The visualization can be applied to any linguistic problem studying an interaction between sever...

Isis

This essay describes essential considerations and select methods in computational text analysis for use in the study of history, specifically the history of science and biomedicine. It explores specific approaches that can be used for understanding conceptual change over time in a large corpus of documents. By way of example, using a corpus of 27,977 articles collected on the microbiome, the essay studies the general microbiome discourse for the years from 2001 to 2010, examines the usage and the sense of the word "human" from 2001 to 2010, and highlights shifts in the microbiome discourse from 2001 to 2010.