Finding Poetry Amid Historic News Pages
University Communications
Author
11/02/2015
Added
72
Plays
Description
The difficulty of culling millions of poems from historic newspapers has left a gap in this important aspect of historical research. To recover the poetry, a collaboration between UNL Libraries and computer science is developing a unique indexing and retrieval method based on visual cues rather than text. The technique may open new possibilities in searching for interesting patterns in other large datasets.
Searchable Transcript
Toggle between list and paragraph view.
- [00:00:00.903]And then this newspaper is from 1876.
- [00:00:03.607]And the poem is here.
- [00:00:05.160]Poetry often shared the front page
- [00:00:07.678]with headlines of the day
- [00:00:09.018]in 19th century newspapers.
- [00:00:10.950]Political candidates
- [00:00:12.268]might be topics of poems.
- [00:00:13.753]The eye quickly zeroes in
- [00:00:15.924]on a poem.
- [00:00:16.991]I'm sort of drawn to it
- [00:00:18.215]because of the white space
- [00:00:19.078]and the jagged features
- [00:00:20.646]of both sides of the poem.
- [00:00:22.031]Researchers
- [00:00:23.100]at the University of Nebraska-Lincoln
- [00:00:24.951]are looking for an efficient way
- [00:00:26.603]to spot poetry
- [00:00:27.656]in the millions of pages
- [00:00:29.257]that have been digitized
- [00:00:30.709]as part of Chronicling America,
- [00:00:32.831]a database of historic newspapers.
- [00:00:35.812]You've got the material there.
- [00:00:37.051]If you can't find
- [00:00:37.955]what you're ultimately interested in,
- [00:00:39.040]it's not that useful.
- [00:00:40.756]And so we want to be pushing people
- [00:00:43.158]to be thinking more broadly
- [00:00:44.745]about how we access
- [00:00:45.912]and how we use these collections.
- [00:00:47.849]The answer
- [00:00:48.683]is a collaboration
- [00:00:49.834]between UNL libraries
- [00:00:51.523]and computer science.
- [00:00:53.428]For computer students
- [00:00:54.741]capture this process,
- [00:00:56.260]this vision process,
- [00:00:57.745]is not easy.
- [00:00:58.479]We learned that
- [00:01:02.496]archived digitized newspaper pages,
- [00:01:06.291]they come in all shapes and sizes.
- [00:01:09.677]Some are so noisy.
- [00:01:11.196]Some are so poorly maintained.
- [00:01:14.335]So one size doesn't fit all.
- [00:01:17.272]A software program
- [00:01:18.923]teaches the computer
- [00:01:20.158]to search for poetry
- [00:01:21.443]using images instead of text.
- [00:01:24.080]Students who are part of the project
- [00:01:26.367]learn communication, team work,
- [00:01:28.423]and problem-solving skills.
- [00:01:30.306]There's no precedence set
- [00:01:32.357]for what we're doing right now.
- [00:01:36.246]So we're basically having
- [00:01:37.332]to start from scratch
- [00:01:38.133]and deal with each bug and problem
- [00:01:40.169]as it comes up.
- [00:01:41.372]In a research setting,
- [00:01:42.488]you don't necessarily know
- [00:01:43.822]there's a way to do it.
- [00:01:44.892]You're trying to get it done
- [00:01:47.244]and see if it works.
- [00:01:49.298]It's more open-ended.
- [00:01:50.983]The technique could be used
- [00:01:52.669]for other large data sets
- [00:01:54.337]and even information
- [00:01:55.689]in other languages.
- [00:01:57.142]If poems have these characteristics
- [00:01:59.796]regardless of language,
- [00:02:00.746]let's say,
- [00:02:01.914]so that means we don't have
- [00:02:03.416]to understand the language
- [00:02:04.652]in order to detect the poems.
- [00:02:06.673]So this is not just for English.
- [00:02:08.756]Funded
- [00:02:09.709]by the National Endowment
- [00:02:10.608]for the Humanities,
- [00:02:11.778]researchers plan to create a database of poems
- [00:02:14.849]linked to Chronicling America.
- [00:02:16.683]UNL's, of course, a world leader
- [00:02:18.836]in digital humanities.
- [00:02:20.205]This is very much the place
- [00:02:21.239]to be doing this sort of research.
- [00:02:22.875]Making sense
- [00:02:24.228]of the digital world
- [00:02:25.545]helps scholars study the past.
The screen size you are trying to search captions on is too small!
You can always jump over to MediaHub and check it out there.
Log in to post comments
Embed
Copy the following code into your page
HTML
<div style="padding-top: 56.25%; overflow: hidden; position:relative; -webkit-box-flex: 1; flex-grow: 1;"> <iframe style="bottom: 0; left: 0; position: absolute; right: 0; top: 0; border: 0; height: 100%; width: 100%;" src="https://mediahub.unl.edu/media/4642?format=iframe&autoplay=0" title="Video Player: Finding Poetry Amid Historic News Pages " allowfullscreen ></iframe> </div>
Comments
0 Comments