Text analysis from…• documents down to words (“low-level”)• words up to documents (“high-level”)
Using documents to learn aboutlanguage(or other social phenomena)Analyzing documents as records/proxies oflanguage, social structures, events, etc.Linguistic studies:morphology, word counts, syntax, etc. … over time (e.g., Google ngram viewer)language across corpora (e.g., politicalspeeches)Underwood, T. (2012). Where to start with text mining.
Using documents to learn aboutlanguage Historical culturomics of pronoun frequencies
Using documents to learn aboutlanguage Universal properties of mythological networks
Using language to learn aboutdocumentsAnalyzing documents as artifacts themselves, withtheir own properties and dynamicsLiterary, documentary studies:Structural/rhetorical/stylistic analysisDocument categorization, classificationDetecting clusters of document features (topicmodeling)Underwood, T. (2012). Where to start with text mining.
Using language to learn aboutdocuments Literary Empires, Mapping Temporal and Spatial Settings in Swinburne
Using language to learn aboutdocuments Using Word Clouds for Topic Modeling Results
What are documents?For this discussion, digital versions of works of spoken or written languageExamples: books, articles, transcripts, emails, tweets…
Documents as contextDocuments have:• form(at)• style• provenance• entities• intentions
Why study documents?• Describe a corpus• Compare/organize documents• Locate relevant information/filter out irrelevant information
Describing a corpus• Finding regularities/differences across groups of documents• Developing theories of structure, style, etc. that can then be tested or applied• May be manual (content analysis) or computer-assisted (statistical)
Differences offormat, genre, participants…• Articles may have sections, but these will vary by discipline and type of article• Books may be fiction or non-fiction (or both)• Transcripts may refer to multiple speakers, non-text content• …ad infinitum
Example: LiteratureFingerprinting Keim, D. A., & Oelke, D. (2007). Literature fingerprinting: A new method for visual literary analysis. In IEEE Symposium on Visual Analytics Science and Technology, VAST 2007 (pp.115-122). doi: 10.1109/VAST.2007.4389004
Organizing documentsDetect similarity between documents and aknown category (or simply amongthemselves)Supports browsing, sentimentanalysis, authorship detection
Example: Bohemian BookshelfThudt, A., Hinrichs, U., & Carpendale, S. (2012). The Bohemian Bookshelf: Supporting Serendipitous BookDiscoveries throughInformation Visualization. In CHI 12: Proceedings of the SIGCHI Conference on Human Factors in ComputingSystems, to appear.
Similarity based on…• common document attributes authorship, genre• common language patterns topics, phrases• common entity references characters, citations
Example: QuantitativeFormalismAllison, S., Heuser, R., Jockers, M., Moretti, F., & Witmore, M. (2011). Quantitative formalism: Anexperiment. Pamphlets of the Stanford Literary Lab (vol. 1).
Classification• assigning an object to a single class• often supervised, using an existing classification scheme and a tagged corpus
Example: Relative signaturesJankowska, M., Keselj, V., & Milios, E. (2012). Relative n-gram signatures: Document visualization at the levelof character n-grams. In Proceedings of IEEE Conference on Visual Analytics Science and Technology 2012(pp. 103-112).
Categorization• assigning documents to one or more categories• suggestive of unsupervised clustering techniques• design choices made to fit particular tasks or goals
Example: UCSD Map ofScienceBörner, K., Klavans, R., Patek, M., Zoss, A. M., Biberstine, J. R., Light, R. P., Larivière, V., &Boyack, K. W. (2012). Design and update of a classification system: The UCSD Map of Science. PLoSONE, 7(7), e39464.
Text is only one component of a document.Research questions often push us to becreative with how we operationalizeconstructs.The richness of language and documents isbest preserved by usingmultiple, complementary approaches.