This repository contains a few text mining routines written in R using a slightly modified version of the RTextTools package. Ultimately, it is hoped that an example script can be provided that shows how these routines can be applied to metadata from the Macaulay Library sound collection in order to translate free-form recordist comments into standard categorical behavior classifications.

Information for the users of this code:

Aside from using this project as a means of learning the basics of machine learning and text mining, I also used it as an opportunity to improve my competency in the R language. Writing this code took a great deal of trial and error, and I found myself referencing many helpful online resoures when I encountered conceptual challenges or errors that I did not understand. The vast majority of the time I spent online I spent on StackOverflow (stackoverflow.com), although I also referenced a number of other sources, including inside-R (inside-r.org) and the personal blogs of R users. In most cases, I tried to learn from the sources I found rather than copying code that I found online verbatim. There were a few functions which I discovered while doing my research that suited needs particularly well, and these I included in my code with attribution to the content creators. However, there are many cases there are only a few ways of doing something in code, and as a result some of my code may resemble code used in particular online examples that are not cited here, whether or not I ever looked at them. If, while using this code, you happen to notice any section of code that is sufficiently similar to a particular online example, to warrant a specific citation, especially if it resembles code you yourself have written, please let me know as soon as possible. I will determine whether I looked at that example at any point, and if I did, I will edit my code to add an attribution to the example in question. It is not my goal to plagiarize from anyone, and I will happily address any missing citations as soon as I am made aware of it. Thank you.

List of all code citations:

A. Liaw and M. Wiener (2002). Classification and Regression by randomForest. R News 2(3), 18--22.

Andrea Peters and Torsten Hothorn (2015). ipred: Improved Predictors. R package version 0.9-4. http://CRAN.R-project.org/package=ipred

Brian Ripley (2015). tree: Classification and Regression Trees. R package version 1.0-35. http://CRAN.R-project.org/package=tree

Christian Buchta, Kurt Hornik, Ingo Feinerer and David Meyer (2015). tau: Text Analysis Utilities. R package version 0.0-18. http://CRAN.R-project.org/package=tau

David Meyer, Evgenia Dimitriadou, Kurt Hornik, Andreas Weingessel and Friedrich Leisch (2014). e1071: Misc Functions of the Department of Statistics (e1071), TU Wien. R package version 1.6-4. http://CRAN.R-project.org/package=e1071

Dirk Eddelbuettel and Romain Francois (2011). Rcpp: Seamless R and C++ Integration. Journal of Statistical Software, 40(8), 1-18. URL http://www.jstatsoft.org/v40/i08/.

Godbole, S., & Sarawagi, S. (2004). Discriminative methods for multi-labeled classification. In Advances in Knowledge Discovery and Data Mining (pp. 22-30). Springer Berlin Heidelberg.

Hadley Wickham (2011). The Split-Apply-Combine Strategy for Data Analysis. Journal of Statistical Software, 40(1), 1-29. URL http://www.jstatsoft.org/v40/i01/.

Ingo Feinerer, Kurt Hornik, and David Meyer (2008). Text Mining Infrastructure in R. Journal of Statistical Software 25(5): 1-54. URL: http://www.jstatsoft.org/v25/i05/.

Jarek Tuszynski (2014). caTools: Tools: moving window statistics, GIF, Base64, ROC AUC, etc.. R package version 1.17.1. http://CRAN.R-project.org/package=caTools

Jerome Friedman, Trevor Hastie, Robert Tibshirani (2010). Regularization Paths for Generalized Linear Models via Coordinate Descent. Journal of Statistical Software, 33(1), 1-22. URL http://www.jstatsoft.org/v33/i01/.

R Core Team (2015). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-project.org/.

Revolution Analytics and Steve Weston (2014). doParallel: Foreach parallel adaptor for the parallel package. R package version 1.0.8. http://CRAN.R-project.org/package=doParallel

Rinker, T. W. (2013). qdap: Quantitative Discourse Analysis Package. version 2.2.0. University at Buffalo. Buffalo, New York. http://github.com/trinker/qdap

Timothy P. Jurka, Loren Collingwood, Amber E. Boydstun, Emiliano Grossman and Wouter van Atteveldt (2014). RTextTools: Automatic Text Classification via Supervised Learning. R package version 1.4.2. http://CRAN.R-project.org/package=RTextTools

Timothy P. Jurka and Yoshimasa Tsuruoka (2013). maxent: Low-memory Multinomial Logistic Regression with Support for Text Classification. R package version 1.3.3.1. http://CRAN.R-project.org/package=maxent

Venables, W. N. & Ripley, B. D. (2002) Modern Applied Statistics with S. Fourth Edition. Springer, New York. ISBN 0-387-95457-0

Xavier Robin, Natacha Turck, Alexandre Hainard, Natalia Tiberti, Frédérique Lisacek, Jean-Charles Sanchez and Markus Müller (2011). pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics, 12, p. 77. DOI: 10.1186/1471-2105-12-77 http://www.biomedcentral.com/1471-2105/12/77/

The metadata that was used to test these routines comes from the Macaulay Library, a large collection of sound and audio files maintained by the Cornell Lab of Ornithology (http://macaulaylibrary.org/about). It is hoped that these data will eventually be made available here, along with a version of the provided script that runs analyses on them. Although the CSV file containing the data used in this script is not currently available to the public, each entry in said CSV file comes from an audio file that may be accessed on the Macaulay Library website by any internet user. Although the Macaulay Library hosts video media in addition to sounds, none of the entries in this dataset pertain to video.