-
If you are citizen of an European Union member nation, you may not use this service unless you are at least 16 years old.
|
Meeting (2016-04-15)
Page history
last edited
by Alan Liu 8 years, 6 months ago
Preliminaries
Quick Orientation & Review of Project Work (update reports indicated below to follow at end of meeting)
(The orientation is for the benefit of newcomers. Updates concern recent project work.)
- Overview
- Project goals
- Stage 1 Goals
- Scrape articles and documents in public discourse mentioning "humanities," "liberal arts," and (in Commonwealth nations) "the arts."
- Current corpus:
- ~30K plain-text documents from: NY Times, Wall St. Journal, LA Times, Washington Post, NPR, The Guardian (London), New Yorker.
- metadata and raw scrapes in spreadsheets (awaiting further refinement)
- storage in Google Drive (awaiting development of MongoDB and Manifest Schema)
- Stage 2 Goals
- Current project work: Topic model the "ambient" public discourse in which the humanities participate.
- Experiment with creating topic models
- Implement topic-model exploration interface. (See our research on interfaces)
- Develop protocol for combined machine/human interpretation of topic models (e.g., cluster analyses + human panels) -- See draft protocol.
- Future project work: Identify articles in the corpus that are centrally "about" the humanities and topic model those.
- Optional future work: perform other kinds of text analysis on WE1S corpus (e.g., use of Cambridge Concept Lab's "Shared Lexis Tool", word embedding, etc.) (see Ben Schmidt, "Word Embeddings for the Digital Humanities")
- Stage 3 Goals:
- Ultimate (Applied) Goals:
- Advocacy kit
- Scholarship on the humanities and on digital-humanities methods
- Additional Methodological Goals:
- Provenance system for DH ("Manifest Schema" -- Overview)
- Current ongoing project work:
- Parallel development cycle:
- Manifest system and backend MongoDB system --updates?
- Other ongoing development work:
- Nathalie Popa: scraping Globe & Mail -- updates?
- Austin Yack: "What Politicians Have to Say About the Humanities" and supporting spreadsheets for his research on the U.S. White House, Congress, and California State Legislature (using Google Visualization API to output HTML tables from Google Spreadsheets. See experiment.)
"Big Picture" of Next Year Work (beginning Summer 2016)
Alan will use his just-submitted internal UCSB grant proposal for the third year of the project as a framework for forecasting and planning of next academic year's work:
- Alan's Academic Senate grant proposal for 2016-17: (excerpts]) --
- "This third request for an Academic Senate research grant (which I anticipate will be the last) is focused on bringing "WhatEvery1Says" forward to the point where it can disseminate research results based on its existing corpus of collected materials; submit a collaborative proposal for the major annual refereed conference in the digital humanities field; and submit a proposal for an extramural NEH digital-humanities Digital Humanities Advancement Grant (DHAG), Level 3 (and possibly additional proposals to such agencies as the ACLS]. In other words, this proposal is for funding to allow the project to "graduate" to a national stage of research and development supported by other sources.... Plans for 2016-17 are chiefly to:"
- Finalize the project's current corpus of public documents, and create publicly-viewable metadata for that corpus.
- Disseminate our topic models and analyses in a public-facing site designed to allow for interactive exploration of the topics, their interrelationships, and sources.
- Submit an extramural grant proposal to the National Endowment for the Humanities (Office of Digital Humanities) to scale up coverage of source publications and nations. In particular, we are exploring the possibility of a Digital Humanities Advancement Grant (DHAG), Level 3 (up to $325,000).1 We have discussed "WhatEvery1Says" in a preliminary way with a grant officer at the NEH Office of Digital Humanities, who has informally indicated the agency's interest. If we can bring "WhatEvery1Says" forward to the stage where it is feasible to apply for the DHAG grant, we may especially link our proposal to the NEH's new initiative titled "The Common Good: The Humanities in the Public Square." Note: for the purposes of the NEH DHAG Level 3 grant, work on "WhatEvery1Says" conducted to date on UCSB Academic Senate grants counts as the prerequisite "start up" phase. (Additionally, we may submit extramural proposals for grants related to the digital humanities from the American Council of Learned Societies--e.g., for the ACLS Digital Extension grants or ACLS Digital Innovation grants.)
- Submit a collaborative paper and panel proposal for Digital Humanities 2017.
- Submit a collaborative article for DHCommons Journal, which publishes peer-reviewed articles on in-progress digital-humanities projects.
- Evolve further a virtualized workstation environment for our topic modeling work and also finalize our Manifest Schema provenance system and its backend MongoDB platform.
- The following is an estimated breakdown of hours by task :
- Assist in finalizing the "WhatEvery1Says" document corpus, and preparing metadata for the corpus: 140 RA hours (Summer 2016)
- Assist in topic modeling the final corpus and analyzing the topic models (through a series of iterative topic models and interpretation steps): 140 RA hours (Summer-Fall 2016)
- Assist in preparing a grant proposal for the National Endowment for the Humanities Digital Humanities Advancement Grant, Level 3 (and possibly also other grants from the NEH or ACLS): 80 RA hour (twice-a-year dealines: "first deadline will be January 11, 2017. The second will be six months later [July 2017])
- Assist in preparing submission of papers and of a panel proposal for Digital Humanities 2017 and also of an article for DHCommons Journal: 20 RA hours. (Note: in addition to assisting with preparing submission materials and conducting the submission process, graduate student RAs may electively be co-authors of papers depending on the relation of their project research to the specific paper, their role in co-writing the paper, etc. I have good results in past grant-driven projects incubating graduate-student co-authored publications and hope to reproduce that as a result of "WhatEvery1Says.") (DH 2017 panel submission deadline: Nov 1, 2016?)
- Assist in technical work on the project's digital front-end and back-end: 140 RA hours (MAT students if available). Specific tasks include:
- creating a public-facing interactive site for exploring topic models (based on adapting and customizing existing systems for this purpose such as DFR-Browser).
- finalizing virtualized/containerized workstation environments for the project's topic modeling and analysis work.
- finalizing the Manifest Schema system of provenance forms
- finalizing the project's MongoDB database to support querying/dynamic generation of metadata about the project's corpus and topic models; and also querying/dynamic generation of provenance metadata.
Discussion
- Partnering (how/when partner institutions are interested in participating)
Meeting (2016-04-15)
|
Tip: To turn text into a link, highlight the text, then click on a page or file from the list above.
|
|
|
|
|
Comments (0)
You don't have permission to comment on this page.