Text Analyzer - September 2017

Feedback, Comments, Mentions, Questions on Twitter, Email, and Facebook

  1. "not effective in identifying relevant research"

    --email from psychology PhD in Illinois, Sept 2 2:34
  2. "Thanks for the feedback! We’re very sorry to hear that Text Analyzer wasn’t effective for you. The tool is still in beta and we are working actively to improve it. Would you be willing to tell us more about what you searched (if you uploaded a document, would you be willing to share that with us?) and what you were looking for? Digging into the details just a bit might help us figure out how to improve the tool for yourself and others."

    --reply to psychology PhD from Alex, Sept 2 10:23
  3. "Thank you for your response, no need to apologize, I think it's an ingenious idea which is why I wanted to provide feedback but then found myself struggling to articulate the issue. I have attached a couple of the articles I think I tried out. I am working on a literature review that encompasses research relevant to the utilization of biomarkers, specifically pupil dilation, stimulated by emotionally evoking words or photos in trying to identify distinguishing features between individuals with PTSD and those who may be malingering. Along with that I was trying to review relevant methods and measures that have proven effective and valid in previous research that. As instructed I tried typing in my own phrases in to the prioritized terms, selecting additions from the identified terms and adjusting the priority of the terms selected but didn't get very far. Perhaps I was missing something but my experience was that these actions did little in the way of narrowing my search or improving the relevance of the articles. I hope this feedback is helpful, I think this idea has great potential! Please let me know if I can provide any further insight."

    --reply to Alex from psychology PhD, Sept 2 13:52
  4. "Wow, thanks you SO much for sharing all this detail -- it's incredibly helpful. We've run these three documents and can see what you're talking about. As it happens, we're working on a new algorithm right now which we think might help in instances like yours. (It will of course still be limited by what material exists in JSTOR). Would you mind if I got in touch with you when the new recommendation engine is complete to see how we did?

    Best of luck in your important research!"

    --reply to psychology PhD from Alex, Sept 3 10:54
  5. Have you tried our new search tool, Text Analyzer (beta) yet? Drop your doc and we show you articles related to it! https://t.co/EmGHKk7f7j https://t.co/rC37Oabzju
    Have you tried our new search tool, Text Analyzer (beta) yet? Drop your doc and we show you articles related to it!  http://jstor.info/c8TF30f1ggT  pic.twitter.com/rC37Oabzju
  6. This is fabulous!!! It would be so easy to drop articles and papers I've save from Dropbox in here.

    -J.E., via Facebook, Sept 11 15:02
  7. Could you post reminders of this feature, say, once a week? Then when I'm back at uni I'll definitely remember to use it

    😂


    -A.C., via Facebook, Sept 11 15:15
  8. "This is really nice...Can you be able to export it inform of a reading list so I can read all content related to a specific term? Example export articles links on Cryptocurrency!"

    --email from economics professor in Kenya, Sept 11 15:30
  9. "Dear [redacted], thanks for the kind words, I’m so pleased that Text Analyzer is helpful to you. There’s no way currently to export the entire article list, but you *can* generate a list from results (which you can then export, or print, or whatever you want. For each result that you’re interested in, just click on the “Add to My Lists” button. You’ll need to create a (free) myjstor account, but once you do this, you can create any number of lists and maintain them to help in your research.

    I hope this is helpful. Do let me know if you encounter any issues!"

    --reply to economics professor from Alex, Sept 11 15:47
  10. @AFairAffair oh, it's real alright. (Tell us what you think? we keep trying to make it better).
  11. This is a REALLY cool way to search JSTOR. Off campus, go to the regular JSTOR via the library page & there's a link.  https://twitter.com/JSTOR/status/907318048721424384 
  12. @SCSULibrary Thanks for sharing! Here's a LibGuide that covers Text Analyzer & other tools on JSTOR.  http://jstor.info/9vlw30f4Ux9 
  13. this is 190% still my favourite thing on the world wide web

    -C.M, via Facebook, Sept 12 6:29
  14. "I tried that one too I was doing from my School/Institution, but when I click add to my list, I am redirected to a login page( I log in using my credentials without any error), on searching the article, I am told I don't have any rights to the article. I have now resulted to downloading the PDF's"

    --reply to Alex from economics professor, Sept 12 9:25
  15. Dice @JSTOR : si subis tu investigación a su Text Analyzer, el sistema busca los docs más pertinentes ¿lo probaste?  http://www.jstor.org/analyze/ 
  16. "This a pretty neat tool, I was wondering if there's an API for it? I couldn't find it on your website hence this e-mail. Cheers!"

    --email from developer in India, Sept 15 8:53
  17. "Thanks for the kind words! We don’t currently have a public API for the tool, but have been considering one. It would help as we design it to understand how you’d want to use it: would you be willing to tell us a bit more about what you’d like to use it for? "

    --reply to Indian developer from Alex, Sept 15 9:27
  18. "I'd like to make this more "real-time", think related videos on YouTube: suggest articles to researchers as they're writing.

    Text Analyzer is great, but overtime a content of a paper changes and it'll be cumbersome to reupload it everytime you want to find relevant material. Isn't it?

    Is it fine for you if I crawl the results (very few hits I promise) to mock up a few things? Let me know.

    Product-wise, for now I can tell you that the topical break-down is a bit TMI. I'd rather see the articles first and the break-down later."

    --reply to Alex from Indian developer, Sept 15 9:49
  19. "Thanks [redacted], this is super-helpful information.

    What do you write your papers in? Word? Google Docs? Overleaf? Etc. We’ve been toying with a Chrome extension that does this updating in real time, which would work with Docs, but not some of the other authoring environments.

    Sure, if you’re just mocking a few things up, go ahead and crawl a few times."

    --reply to Indian developer from Alex, Sept 15 9:58
  20. <blushing> You're close -- actually, and even better, IT'S SCIENCE! (Tell us how it works for you! We at @JSTOR Labs are #alwayslearning.)  https://twitter.com/MisterBDaniel/status/908712091099521025 
  21. Take a look at JSTOR's (beta) tool - Test Analyzer! Use your own document to search for articles and books. https://t.co/75g7bXX2Cw https://t.co/AR03dtzwkM
    Take a look at JSTOR's (beta) tool - Test Analyzer! Use your own document to search for articles and books.  http://ow.ly/f7uz30fbuC7  pic.twitter.com/AR03dtzwkM
  22. "Well, a Chrome extension is cool but in my experience in working with researchers they don't use Google Docs to author their papers since its terrible for long-form work.

    Also, to give you a clearer idea of the intended end-result take a look at the demo of our product.

    At the end there is a screen called "Plagiarism Stack", I want to create a similar page in which users can see all articles relevant to what they're working on. I think it'll blow peoples' minds :P. "

    --reply to Alex from Indian developer, Sept 18 3:19
  23. "Thanks, [redacted], that’s very helpful. Okay, so what you’d need is an API that allows you to pass an arbitrarily large amount of text and it would return a set of links to articles in JSTOR. I’ll reach out to you when that API becomes public.

    Cool product, by the way!"

    --reply to Indian developer from Alex, Sept 18 10:05
  24. "Yup, precisely that. If you can return the journal name etc in the output that would be cool as well. Any idea of the time frame to release? If you want Beta testers for the API we can be involved as well.

    Aw, thanks. "

    --reply to Alex from Indian developer, Sept 18 10:47
  25. Stuck in a research rut? Try searching with Text Analyzer (beta)! Upload your doc and find related articles: https://t.co/Jcz6pP3I7D https://t.co/5TsGPCyiHC
    Stuck in a research rut? Try searching with Text Analyzer (beta)! Upload your doc and find related articles:  http://jstor.info/89q130ff2rU  pic.twitter.com/5TsGPCyiHC
  26. "We’re sorry to hear that Text Analyzer didn’t work for you! We’d love to work with you to figure out what happened and hopefully get it working. To start, I have a few questions if you’re able to answer so we can troubleshoot from our end.

    You mentioned you were using an iPad and taking a photo of a document to then be analyzed:
    Can you tell me what version of iOS you are using on the iPad?
    Can you also send the document you were taking a photo of? I can try printing it off here and taking a photo with our iPad to see if I encounter the same issue.
    Lastly, which browser on the iPad were you using when taking a photo?

    I appreciate your help and hope to hear from you soon!"

    --email to HS librarian in UK from Aparna, after bug report, Sept 18 14:48
  27. "Text Analyzer from JSTOR Labs lets you upload a document and search for articles based on your text!"  https://twitter.com/JSTOR/status/909842012274388994 
  28. "Thanks for your email. To confirm:

    iPad is on iOS 10.3.3
    Text Analyzer was on Safari browser
    Photo of text (IMG 7113) is attached, but I have used several books to take a photo from, and what used to work every time on any book, now fails every time.

    Text analyzer (TA) works well on my iPhone on iOS 10.3.3 on Safari. However, the mobile (cell) version of TA is different from the iPad desktop version. I attach a picture (JSTOR_TA1) which shows the iPhone version, and you can see it has a “take picture” button. The other picture (JSTOR_TA2) shows the interface on the iPad is different, and what looks to be the desktop site. A photo can still be taken through “select files”.

    Is the problem related to cell v desktop versions, iPhone v iPad?"

    --email to Aparna from UK librarian, Sept 19 4:50
  29. "Thank you for your response! I have a few follow-up questions and comments:

    When you click the ‘Select Files from Computer’ link on Text Analyzer on your iPad, and you take a photo from a book page, where in the analysis is it failing? In the progress page, at which point is it freezing failing at: Uploading your document, Analyzing text and identifying topics orGenerating recommendations?

    Also, I want to be sure you are not expecting the iPad to have the same behavior as a phone where you are given the option to ‘Take a Snapshot’. The iPad will not have that option and will mimic a desktop version of Text Analyzer instead.

    I will try reproducing the issue here with our iPad, and may ask a few more questions if needed.
    I appreciate your patience with this matter, and thank you again for your help in tracking down what could be happening!"

    --email to UK librarian from Aparna, Sept 19 9:09
  30. "...I hope that I will not miss your visit in October. Please let’s keep in touch. I love the Text Analyzer, it is a cool tool."

    --email from library dean in Massachusetts to Outreach, Sept 25 14:03