During class we used the Voyant text mining tool to analyze a set of document that was provided to us. We were Group 6.
What information did you choose to delete and keep? Why? For example, did you choose to delete or keep the table of contents?
The word that was kept had to be different enough to provide different aspects of the total narrative, words that repeated in similar situations, such as (Slave, Slavery, Slaves) served the same or similar roles in the stories told and were large chunks of the total text so they had to be taken into the same umbrella to tell a fuller story.
What stopwords did you choose and why? What was your thinking in terms of how they would impact your work in Voyant and your reading of the text?
about, above, across, after, afterwards, again, against, all, almost, alone, along, already, also, although, always, am, among, amongst, amoungst, amount, an, and, another, any, anyhow, anyone, anything, anyway, anywhere, are, around, as, at, back, be, because, been, before, beforehand, being, beside, besides, between, both, bottom, but, by, call, can, cannot, cant, co, con, could, couldnt, de, did, didn’t, do, does, doesn’t, don’t, done, down, due, during, each, eg, eight, either, eleven, else, elsewhere, enough, etc, even, ever, every, everyone, everything, everywhere, except, few, fifteen, fify, fill, find, fire, first, five, for, former, formerly, forty, found, four, from, front, full, further, get, give, go, got, had, has, hasnt, have, he, hence, her, here, hereafter, hereby, herein, hereupon, hers, herself, him, himself, his, how, however, hundred, ie, if, in, inc, indeed, into, is, it, its, itself, keep, last, latter, latterly, least, less, ltd, made, many, may, me, meanwhile, might, mill, mine, more, moreover, most, mostly, move, much, must, my, myself, name, namely, neither, never, nevertheless, next, nine, no, nobody, none, noone, nor, not, nothing, now, nowhere, of, off, often, on, once, one, only, onto, or, other, others, otherwise, our, ours, ourselves, out, over, own, part, per, perhaps, please, put, rather, re, said, same, say, see, seem, seemed, seeming, seems, serious, several, she, should, since, six, sixty, so, some, somehow, someone, something, sometime, sometimes, somewhere, still, such, system, take, ten, than, that, the, thee, their, them, themselves, then, thence, there, thereafter, thereby, therefore, therein, thereupon, these, they, thing, third, this, those, thou, though, three, through, throughout, thru, thus, thy, to, together, too, toward, towards, twelve, twenty, two, un, under, until, up, upon, us, very, via, was, we, well, were, what, whatever, when, whence, whenever, where, whereas, whereby, wherein, whereupon, wherever, whether, which, while, whither, who, whoever, whole, whom, whose, why, will, with, within, without, would, yet, you, your, yours, yourself, yourselves.
These words were both auto-generated and selected for their application to the abilities of them to affect the true narrative being told by over-saturating the prominent words displayed, there are a few which could serve a purpose, but for the way I approached the text analysis, I found they served a less critical role.
What Voyant tools did you find most helpful and relevant? Why? (Discuss at least three tools.)
The context tools were the most helpful because they filled in the blanks in the narratives that consistently occurred without having to track down the individual phrase, i.e., “it would have been considered impudent for any of the passengers to have spoken to her, and the crew were not allow to have any conversation with them. When we reached St. Louis, the slaves were removed to a boat bound for New Orleans, and the history of the beautiful slave -girl remained a mystery. I remained on the boat during the season, and was not an unfrequent occurrence to have on board gangs of slaves on their way to the cotton, sugar and rice plantation of the south. Toward the latter part of the summer Captain Reynolds left”. Another tool that was helpful was Categories. With the Categories tool we were able to group words together that meant similar things. This helped up in our exploration of other pieces of text that may be impactful too. The last tool that I found to be useful is the Trends tool. With the Trends tool we are able to see a line graph of words that we have chosen. This graph displays how frequent a certain word (or list of words) occurs in a piece of text. It came in handy as we could pinpoint certain key ideas between each text by viewing how much a certain term was present.
What struck you most about what Voyant did or didn’t reveal? (Where you expecting to see something and didn’t? Was it what you expected? Etc.)
It was incredible how Voyant could find repeated phrases that fit together in similar uses versus phrases that may seem to serve similar purposes but were used to describe a very different situation.
Based on what you see in Voyant, how might you read the text differently in the future, e.g., what else might you read or look for?
After using Voyant I believe that analyzing how frequent a writer uses a certain set of words can help me in better understanding the text that I am reading. Additionally, if the writer uses the same word or phrase in different contexts it can have significance to the central idea of the document and can lead to the creation of better questions surrounding the topic.
Be critical of the tool. What are some of Voyant’s potential flaws and pitfalls? How might it be “misused”?
In my experience, the Voyant tool is not very robust. For instance, during class we had a hard time getting all the documents to load since the site may not be able to handle many users at once. Some of the features within the tool were also buggy. When trying to create Categories it would take a few attempts before the new Categories would be added to my text.
While text mining tools like Voyant are great for educational research, they can be misused by people looking to shape false narratives. If you incorporate bias into your text mining efforts then you will come up with results that can mislead people.
What does this approach reveal that other methods do not? Are there limitations to text mining?
Text mining is a good tool for formulating better questions about the texts and topics that you are researching. It can also reveal overlooked notions within these texts that would be difficult to find if you wanted to analyze large amounts of text. It removes a lot of the work in finding correlations between text using software.
Some limitations of text mining involve the work involved in getting the text into tools like Voyant. Before the text can go into these tools it must be “cleaned” which means that you need to remove unnecessary parts such as punctuation. This can take a significant amount of time if the texts are large. Another limitation is that while text mining tools can find similarity between texts, if all the texts are ambiguous or incomplete then it negatively affects the importance of the text mining results.
What impact did your experience with Voyant have on your past reading and understanding of the text?
After putting the following documents into Voyant, it produced these graphs. it impacted me by providing a visual representation of statistics regarding gun control. It helped me understand the statistics and really see the extent of the numbers.