Computational Folkloristics
Call for Papers
Special Issue of the Journal of American Folklore edited by Timothy R. Tangherlini

Over the course of the past decade, a revolution has occurred in the materials available for the study of folklore. The scope of digital archives of traditional expressive forms has exploded, and the magnitude of machine-readable materials available for consideration has increased by many orders of magnitude. Many national archives have made significant efforts to make their archival resources machine-readable, while other smaller initiatives have focused on the digitization of archival resources related to smaller regions, a single collector, or a single genre. Simultaneously, the explosive growth in social media, web logs (blogs), and other Internet resources have made previously hard to access forms of traditional expressive culture accessible at a scale so large that it is hard to fathom. These developments, coupled to the development of algorithmic approaches to the analysis of large, unstructured data and new methods for the visualization of the relationships discovered by these algorithmic approaches - from mapping to 3-D embedding, from time-lines to navigable visualizations - offer folklorists new opportunities for the analysis of traditional expressive forms. We label approaches to the study of folklore that leverage the power of these algorithmic approaches "Computational Folkloristics" (Abello, Broadwell, Tangherlini 2012).

The Journal of American Folklore invites papers for consideration for inclusion in a special issue of the journal edited by Timothy Tangherlini that focuses on "Computational Folkloristics." The goal of the special issue is to reveal how computational methods can augment the study of folklore, and propose methods that can extend the traditional reach of the discipline. To avoid confusion, we term those approaches "computational" that make use of algorithmic methods to assist in the interpretation of relationships or structures in the underlying data. Consequently, "Computational Folkloristics" is distinct from Digital Folklore in the application of computation to a digital representation of a corpus.

We are particularly interested in papers that focus on: the automatic discovery of narrative structure; challenges in Natural Language Processing (NLP) related to unlabeled, multilingual data including named entity detection and resolution; topic modeling and other methods that explore latent semantic aspects of a folklore corpus; the alignment of folklore data with external historical datasets such as census records; GIS applications and methods; network analysis methods for the study of, among other things, propagation, community detection and influence; rapid classification of unlabeled folklore data; search and discovery on and across folklore corpora; modeling of folklore processes; automatic labeling of performance phenomena in visual data; automatic classification of audio performances. Other novel approaches to the study of folklore that make use of algorithmic approaches will also be considered.

A significant challenge of this special issue is to address these issues in a manner that is directly relevant to the community of folklorists (as opposed to computer scientists). Articles should be written in such a way that the argument and methods are accessible and understandable for an audience expert in folklore but not expert in computer science or applied mathematics. To that end, we encourage team submissions that bridge the gap between these disciplines. If you are in doubt about whether your approach or your target domain is appropriate for consideration in this special issue, please email the issue editor, Timothy Tangherlini at tango@humnet.ucla.edu, using the subject line "Computational Folkloristics query". Deadline for all queries is April 1, 2013.

All papers must conform to the Journal of American Folklore's style sheet for authors. The guidelines for article submission are as follows: Essay manuscripts should be no more than 10,000 words in length, including abstract, notes, and bibliography. The article must begin with a 50- to 75-word abstract that summarizes the essential points and findings of the article. Whenever possible, authors should submit two copies of their manuscripts by email attachment to the editor of the special issue at: tango@humnet.ucla.edu. The first copy should be sent in Microsoft Word or Rich Text Format (rtf) and should include the author's name. Figures should not be included in this document, but 'call outs' should be used to designate where figures should be placed (e.g., (insert Figure 1 here)). A list at the end of the article (placed after the bibliography) should detail the figures to be included, along with their captions. The second copy of the manuscript should be sent in Portable Document Format (pdf). This version should not include the author's name or any references within the text that would identify the author to the manuscript reviewers. Passages that would identify the author can be marked in the following manner to indicate excised words: (****). Figures should be embedded in this version just as they would ideally be placed in the published text. Possible supplementary materials (e.g., additional photographs, sound files, video footage, etc.) that might accompany the article in its online version should be described in a cover letter addressed to the editor. An advisory board for the special issue consisting of folklorists and computer scientists will initially consider all papers. Once accepted for the special issue, all articles will be subject to the standard refereeing procedure for the journal. Deadline for submissions for consideration is June 15, 2013. Initial decisions will be made by August 1, 2013. Final decisions will be made by October 1, 2013. We expect the issue to appear in 2014.