Distant Reader Study Carrel

About your study carrel

This page outlines the breadth & depth of your "study carrel" -- the results & analysis of your Distant Reader submission. Peruse the content of this page, and then consider learning how to dig deeper by reading the Distant Reader Study Carrel Cookbook. If you want "just the facts", then consider reading this text's synopsis.

Size & scope

First, the simple things. Your study carrel was created through the submission of a [SINGLE URL|FILE OF URLS|FILE FROM YOUR COMPUTER|ZIP FILE]. This ultimately resulted in a collection of 82 item(s). The original versions of these items have been saved in a cache, and each of them have been transformed & saved as a set of plain text files. All of the following analysis has been done against these plain text files.

Your study carrel is 91117 words long. [0] Each item in your study carrel is, on average, 1111.0 words long. [1] If you dig deeper, then you might want to save yourself some time by reading a shorter item. On the other hand, if your desire is for more detail, then you might consider reading a longer item. The following histograms and box plots illustrate the overall size of your study carrel.


On a scale from 0 to 100, where 0 is very difficult and 100 is very easy, your documents have an average readability score of 43.0. [2] Consequently, if you want to read something more simplistic, then consider a document with a higher score. If you want something more specialized, then consider something with a lower score. The following histograms and box plots illustrate the overall readability of your study carrel.

Word frequencies

By merely counting & tabulating the frequency of individual words or phrases, you can begin to get an understanding of your carrel's "aboutness". Excluding "stop words", some of the more frequent words include: data, digital, library, university, project, work, research, information, metadata, libraries, collections, use, will, access, new, services, dlf, one, user, community, also, content, forum, preservation, open. [3] Using the three most frequent words, the three files containing all of those words the most are ./txt/osf-io-4214.txt, ./txt/osf-io-5061.txt, and ./txt/osf-io-1026.txt.

The most frequent two-word phrases (bigrams) include: dlf forum, digital library, linked data, digital collections, state university, noun project, task task, data curation, working group, montana state, slide number, best practices, sara mannheimer, digital scholarship, university dlf, young sara, mannheimer jason, clark montana, presentation notes, presenter presentation, digital preservation, special collections, number slide, open access, curation network, and the three file that use all of the three most frequent phrases are ./txt/osf-io-5878.txt ./txt/osf-io-8532.txt, and ./txt/osf-io-610.txt.

While often deemed superficial or sophomoric, rudimentary frequencies and their associated "word clouds" can be quite insightful:




Sets of keywords -- statistically significant words -- can be enumerated by comparing the relative frequency of words with the number of times the words appear in an entire corpus. Some of the most statistically significant keywords in your study carrel include: data, library, university, https, digitization, http, metadata, digital, file, privacy, projects, researcher, service, user, access, blockson, collecting, collection, communication, digitizing, programming, researchers, reuse, software, users. And now word clouds really begin to shine:


Through the use of a concordance -- a keyword-in-context tool, or a "poor man's search engine" -- you can see how words are used in relation to other words. Here is a random sample of concordance entries using the two most significant keyword as input:

esearch approaches for tools they needed data acquisition and improved ways of extract the
upport crossinstitutional collaborations data curation network dcn community collaboration
eneur and library community convener and data society fellow ken varnum senior program man
riment with combining different types of data into a single triple store greater accuracy 
 selected a few examples from the linked data implementations described that have put into
lows for processing not only publication data from cvs but also workflows for processing u
 but had very little reference point for data sharing and archiving requirementsnot gettin
mushed in some way losslessly compressed data is smushed in some way but can be unsmushed 
 the research will explore new trends in data management to help develop best practices fo
m to establish a critical mass of linked data on the semantic web in the process we worked
ld we work on making it easier to submit data ill admit that the math totally makes my hea
tion technology ht hathitrust digital library bentley library the rest of um library cleme
chema sites reportsforms drawings photos data group number item number metadata metadata m
eart of every great university is its library the intellectual hub of campus where collecv
tal library alongside those things for d data we will provide more context about the model
data analysis data files final paper the data appendix readme file documents original data
ministrative interface users dont access data or maps in figgy directly httpsmapsprinceton
steering committee meeting depository library council meeting federal depository library c
ml engineer docpropsthumbnailjpeg linked data in practice the ld p initiative amber billey
shing digital collections and geographic data in descriptive metadata bibliographic data e
stuit from the university of michigan library research and publication are hard to engage 
 read the readme and download any of the data you may want to explore or print d vs d at u
data catalog marc journal kb first layer data can comprise many formats and come from many
indings jamie researcher needs jamie raw data technical metadata including instrument conf
tate library host cunningham memorial library indiana state university host american legio

Topic modeling

Topic modeling is another popular approach to connoting the aboutness of a corpus. If your study carrel could be summed up in a single word, then that word might be data, and ./txt/osf-io-7511.txt is most about that word.

If your study carrel could be summed up in three words ("topics") then those words might be: digital, data, and data. And the respective files would be: ./txt/osf-io-610.txt, ./txt/osf-io-1090.txt, and ./txt/osf-io-1026.txt.

If your study carrel could be summed up in five topics, and each topic were each denoted with three words, then those topics and their most significantly associated files would be:

  1. digital library data - ./txt/osf-io-610.txt
  2. data digital work - ./txt/osf-io-1090.txt
  3. data library research - ./txt/osf-io-6591.txt
  4. assessment https org - ./txt/osf-io-1026.txt
  5. digital 3d university - ./txt/osf-io-2314.txt

Moreover, the totality of the study carrel's aboutness, can be visualized with the following pie chart:

Nouns & verbs

Through an analysis of your study carrel's parts-of-speech, you are able to answer question beyonds aboutness. For example, a list of the most frequent nouns (library, ●, project, university, datum, data, digital, user, research, collection, information, work, metadata, community, service, group, dlf, access, year, tool, time, task, forum, system, content) helps you answer what questions; "What is discussed in this collection?" An enumeration of the lemmatized verbs (be, have, do, use, make, create, ○, work, need, provide, include, find, see, want, develop, get, take, build, share, base, help, link, go, look, ’) helps you learn what actions take place in a text or what the things in the text do. Very frequently, the most common lemmatized verbs are "be", "have", and "do"; the more interesting verbs usually occur further down the list of frequencies:



Proper nouns & pronouns

An extraction of proper nouns (University, Digital, Library, Data, DLF, Project, |, Forum, Metadata, Libraries, Research, State, Services, Slide, Collections, Group, Information, October, Assessment, ○, Librarian, •, Presentation, User, Center) helps you determine the names of people and places in your study carrel. An analysis of personal pronouns (we, it, i, you, they, them, us, me, ​the​, itself, he, she, themselves, ’s, ourselves, him, myself, y, yourself, one, her, http://indigenousdigitalarchive.org, https://wiki.diglib.org/assessment:user/reuse, ya, y’) enables you to answer at least two questions: 1) "What, if any, is the overall gender of my study carrel?", and 2) "To what degree are the texts in my study carrel self-centered versus inclusive?" Below are words cloud of your study carrel's proper & personal pronouns.

proper nouns


Adjectives & adverbs

Learning about a corpus's adjectives (digital, other, more, new, good, many, 3d, different, open, •, available, large, own, specific, first, next, current, important, few, institutional, full, able, great, most, such) and adverbs (not, •, also, so, more, just, up, then, out, here, now, only, very, as, together, often, well, most, always, even, in, still, on, really, rather) helps you answer how questions: "How are things described and how are things done?" An analysis of adjectives and adverbs also points to a corpus's overall sentiment. "In general, is my study carrel positive or negative?"




[0] Once upon a time, a corpus of a million words was deemed large.

[1] To put this into context, the typical scholarly journal article is about [NUMBER] words long, Shakespeare's Hamlet is [NUMBER] words long, and the Bible is [NUMBER] words long.

[2] In this case, a Flesch readability score is being calculated. It is based on things like the number of words in a document, the lengths of the words, the number of sentences, the lengths on the sentences, etc. In general children's stories are have lower Flesch scores while insurance documents and doctoral dissertations have higher scores.

[3] "Stop words" are sometimes called "function words", and they are words which carry little or no meaning. Every language has stop words, and in English they include but are not limited to "the", "a", "an", etc. A single set of stop words has been used through out the analysis of your collection.

[4] Concordances are one of the oldest forms of text mining, first developed in the 13th century to "read" religious documents.

[6] An unsupervised machine learning process, topic modeling is a very popular text mining operation. Assuming that a word is known by the company it keeps, topic modeling identifies sets of keywords denoted by their centrality in the text. Words which are both frequent as well as in close proximity to each other are considered significant.