Syllabus for CURS 4990_ Seretha D. Williams Summer 2015
Course Objective: This course is designed to give undergraduate students the opportunity to complete undergraduate research and/or creative activity in any discipline. The research and/or creativity activity can be outside of the student’s chosen major and can take place on either the Summerville or Health Sciences campus. The student and faculty member should agree upon a weekly and/or semester schedule. Roles and expectations of each member of the project should be outlined as well.
Students will actively engage in an undergraduate research project and/or creative activity. As a result of this process, students should be to:
a) “Generate questions/aims/hypotheses framed within structured guidelines. Anticipate and prepare for ethical, cultural, social, and team (ECST) issues.”
b) “Collect and record self-determined information/data, choosing an appropriate methodology based on structured guidelines.”
c) “Evaluate information/data and the inquiry process using self-determined criteria developed within structured guidelines. Refines others’ processes.”
d) “Organize information/data using self or team determined structures, and manage the processes within supervisor’s parameters.”
e) “Analyzes information/data and synthesizes to fully integrate components, consistent with parameters set. Fill knowledge gaps that are stated by others.”
f) “Use appropriate language and genre to address gaps of self-selected audience. Apply innovatively the knowledge developed to a different context. Probe and specify ECST issues in each relevant context.”*
Other learning objectives can be outlined by the faculty mentor.
This course is designed to be a zero or one credit course. Students can register for this course as a zero credit course if they do not wish to earn academic credit, but would like to receive recognition for their undergraduate research and/or creative activity outside of their required coursework. Reasons for taking this course for no credit could be related to the use of HOPE and/or payment of tuition and fees associated with credits. This course can be repeated.
This course will be on the student’s transcript. It is a pass/fail course. The student can fail the course if they do not complete the agreed upon requirements. Students do have the option to withdraw from the course in accordance with university policies and procedures.
Only students in good standing with the university and their major should enroll in this course.
*(John Willison and Kerry O’Regan, August 2008/October 2013: http://www.rsd.edu.au)
Text Mining and Digital Humanities: Quantitative Analysis of African American Poetry
Kenton Rambsy, coordinator of the Black Literature and Digital Humanities: The Black Book Interactive Project, asks, “How can digital technologies enhance African American literary studies?” Scholars including Rambsy, William Andrews, Julian Chambliss, Jessica Marie Johnson, Jonathan Hagood, and others are using technology to recover lost and under studied texts, to conduct textual analyses that link African American texts to African diasporic and continental texts, and to discover themes and linguistic patterns or commonalities. Not only does a technology such as text mining allow researchers to add data to existing qualitative studies, but it also generates new areas of scholarship because the software can reveal patterns and discontinuities within a body of literature previously unseen. The answer to Rambsy’s question, then, is digital technologies such as text mining add value to African American literary studies by allowing researchers to conduct quantitative analyses that enhance the qualitative analyses of studied and under studied texts. Text mining is one tool scholars in digital humanities can use to unpack themes, symbols, word occurrences, linguistic phrasings, allusions, or other literary devices within the oeuvre of an author or across a broad spectrum of authors.
“Text Mining and Digital Humanities: Quantitative Analysis of African American Poetry” is a research project that uses comparative – quantitative and qualitative- analysis to formulate research questions about African American poetry and provide responses to those questions. The project is a digital humanities project, and, as such, it will be multidisciplinary in its scope. We will produce a database of words collected from the works of Langston Hughes, Margaret Walker, Gwendolyn Brooks, Maya Angelou, and Alice Walker, a report of our findings, and an online tool for the public to access our database.
Objectives and Significance
The objectives of my proposed text mining research are to mine, map, and analyze word patterns and occurrences in the poetry of Langston Hughes, Margaret Walker, Gwendolyn Brooks, Maya Angelou, and Alice Walker. Margaret Walker, Gwendolyn Brooks, Maya Angelou, and Alice Walker acknowledge Hughes’s influence on their writing. We will create a word database for the collected poems of five major American authors and make that database available and searchable for the public. The project is relevant because it will provide a tool for scholars- student and faculty- to produce new scholarship on African American poets, promote the inclusion of African American literature in the emerging field of digital humanities, and involve Georgia Regents University in the national and international conversations concerning digital humanities and text mining.
Students will benefit from conducting text-mining research and contributing to the field of digital humanities. They will learn to use scanning and text (data) mining software, to do comparative analysis, to find literary criticism using GIL and Galileo, to create a database, and to identify recurrent motifs in African American poetry. In addition, they will visit a digital humanities center, the Margaret Walker Center at Jackson State University, and meet with Dr. Robert Luckett, director of the center. Dr. Luckett will explain the process of creating a digital archive and give us access to Margaret Walker files not yet digitized. The research skills the students learn working on this project will be applicable and relevant to other disciplines.
We will use word analysis as one method of proving (or disproving) measurable influence. Voyeur will be the primary software we use for text-mining. However, we will select one additional software program for comparison. The software will count the number of occurrences of words. We will identify words or word groups with high occurrence. Our hypothesis proposes similar linguistic phrases, cultural referents, geographic sites, or historical references indicate an influence of Langston Hughes on the poets. However, quantitative analysis will be insufficient to draw conclusions.
For our second method, we will consult critical scholarship written by scholars who have compared the poets to Langston Hughes and biographical texts in which the authors have articulated such influences.
Admittedly, the methodology is loose for this project. The looseness is intentional. Text-mining is first about knowledge discovery. We want to test and discover the possibilities and limitations of text-mining literary texts.
Faculty Mentor and Student Roles
As the faculty mentor, I will train the students to use text-mining software, to record data, to interpret data, to find scholarly sources in GIL and Galileo, to develop a searchable database, to create a poster for presentation. I will also arrange the trip to Jackson, MS to tour the Margaret Walker Center and meet with the digital humanities director.
Students will scan and upload digital files into text-mining software, record observations, find patterns or anomalies in the data, produce a report and a poster
What is Text Mining?
Text mining is a way to analyze batches of text and gather quantitative information. Literary textual analysis is usually a qualitative process, but with this project, we will count and measure data in an attempt to generate evidence to answer our research questions. You will determine our overarching research question as well as the smaller, more direct questions that will guide our study. The basic steps of text mining are:
Develop a research question
- Locate the data
- Digitize the data
- Clean the data
- Analyze the data
- Visualize the data
What sorts of questions do Digital Humanities researchers ask?
Digital Humanities researchers ask questions about patterns, frequencies, and meanings. Their questions develop out of the types of analysis they need for a project. They may need to analyze for word frequency; word trends; comparisons; identifying names, places, or periods; or sentiment. They may want to look at the data after it has been collected and then develop a research question or hypothesis (known as a grounded theory question).
What are your responsibilities to the project?
At our first meeting, we will discuss the nature of the project, establish roles, and set deadlines.
Week 1 (May 18-22): Kickoff Event. Reading about text mining, digital humanities. Scanning texts and creating digital files.
Week 2 (May 25-29): Learning to use software. Uploading digital files into text mining software.
Week 3 (June 1-5): Data processing. Mapping and analyzing word patterns and occurrences.
Week 4 (June 8-12): Mapping and analyzing word patterns and occurrences. Travel to Jackson, MS to meet with Dr. Robert Luckett.
Week 5 (June 15-19): Identifying literary criticism that compares poets.
Week 6 (June 22-26): Writing report on research and creating poster for presentation.
Readings and Videos: (I will add to this list as needed.)
- AfAm Lit and Digital Humanities http://www.hastac.org/blogs/krambsy/2014/01/21/african-american-literature-and-digital-humanities
- Intro to topic modeling http://journalofdigitalhumanities.org/2-1/topic-modeling-a-basic-introduction-by-megan-r-brett/
A Companion to Digital Humanities http://www.digitalhumanities.org/companion/
- TED Talk on Google Ngram http://www.ted.com/talks/what_we_learned_from_5_million_books?language=en
Tools for text analysis http://www.tapor.ca/?tab=
Quantative Text Analysis Programs http://academic.csuohio.edu/kneuendorf/content/cpuca/qtap.htm
List of Text Analysis Programs http://academic.csuohio.edu/kneuendorf/content/cpuca/qtap.htm
Google Ngram Viewer https://books.google.com/ngrams/
About ngram viewer http://www.culturomics.org/Resources/A-users-guide-to-culturomics
Poem Viewer http://ovii.oerc.ox.ac.uk/PoemVis/index.html
Voyant Tools http://voyant-tools.org/
Voyant Pedagogy Kit http://pedagogy-toolkit.org/tools/VoyantTools.html
Lexomics to clean the data http://wheatoncollege.edu/lexomics/tools/