Skip to Main Content

Lehigh University Libraries - Library Guides

Using AI Tools for Library Research

Using Keenious

Below is documentation that helps you learn features of “Keenious”, a tool that recommends research papers related to a research question, URL, or uploaded paper.

NOTE: The documentation (checked with the vendor in January 2024) about using Keenious below may not reflect ongoing changes in Keenious functionality or the appearance of its features. For up to date information on features, ask a subject librarian, who can then contact the vendor.



For details and also to monitor developments, see the documentation Keenious provides.

First, create an account by going to the following link; look for the sign up link: "Don't have an account? Sign up."  Watch for a confirmation email.

**Make sure you are logged into Keenious when you use it if you want to use the bookmarking feature and otherwise below suggests that you should be in Lehigh's IP range (or  it would appear to follow, have the library version of the VPN on if you are off campus.) On the other hand, if you use the account login, you don't have to be in the IP range or have on the VPN--the login will you access to all the features.  Ensure that you have access to Lehigh's links out to full text when you use Keenious.

[Details from vendor:

People need to log in if they want to use the bookmarks function. Bookmarks only works for logged in users, because that's the way it recognizes you are the same user as earlier and the bookmarks you make in one session (when logged in) are available to you later as well. Currently, it's the only feature that requires having a user and logging in.

The Lehigh University login ensures that the user gets full access to all of Keenious' features and is connected to Lehigh (even if off campus and even if the user is not on an IP address affiliated with Lehigh).

Users who aren't logged in (and are NOT on a Lehigh affiliated IP address) only gets:
- Top 10 Results
- Analyse up to 1000 words
- Highlight up to 100 words
- 1 daily cross-language search

User that aren't logged in (and are NOT on a Lehigh affiliated IP address) will not be able to go to the next result page].



Start by introducing text into the text box or by specifying a URL or uploading a paper. The example below involves putting in a search statement into the text box. This will generate a list of publications.




The words you search in below will be identified in the title and/or abstract of the articles that the search engine recommended.




Clicking on “Filters” brings up the following, from which you can limit by year range of the search results and specify a range of citation counts, defined as the number of citing items. 

The words you search in the bar will be identified in the title and/or abstract of the articles being recommended.




The default in the search results is to a display of articles, but if you click on the "Topics" option you bring up "topics covered in papers that are similar to yours". 


Below shows that topics can be successively added to the search results. One can go back to topics and add or exclude topics.





You can also find articles similar to one of the search results. Click on the result and scroll down to see similar articles.



You can also highlight text, then click Explore. “Keenious will now emphasize what is mentioned in the part you highlighted when it's analyzing your text. The recommendations from Keenious are still kept relevant to the whole of your document!”



After running a search in Keenious, how do you access the full-text of articles?

Let's say you brought up this search result: 


By clicking on the "Check LU Access" button, you can then link out to the following screen, from which you can access full-text. Had you not see the full text link, you could then order the article by using the "ILLiad" link. (Scroll down to see information about ILLiad set up.)




On the upper right of a search result, notice the double apostrophes, which stand for "Cite Publication", and the star, which stands for "Add to Bookmarks".




 “When you upload your document to Keenious, our system will automatically detect the language used in the document. If the language is not English, the text will be translated into English using a state-of-the-art translation engine. The translated text will then be used as the input for our recommendation engine, allowing you to receive recommendations for English research literature regardless of the language used in your document.” 

To get around the limit on the number of characters that can be translated, use the highlighting capability mentioned above.



See documentation below:



"Based on user feedback, which highlighted a need for clear explanations alongside accurate results, together with multiple universities we developed the Result Insights feature. This feature shows you why the Keenious engine ‘thinks’ a certain result is relevant, focusing on factors such as text relevance, publication date and citation count. It's designed to give you a better understanding of our search and recommendation process by explaining why specific results are suggested and how they stand out from the rest."



From vendor:

Fri, Jan 12, 2024 
I spoke with our developers to be able to accurately respond to your question. It proved to be a harder question to answer than I first anticipated. I tested with the biggest PDF I have and it worked: it's an academic book that was 853 pages long, 88,4 MB, and had about 2.2 million characters.
This is the response from our developers:
Our limit is somewhere in the millions of characters. There is not so much a limit on our end, but there is a potential limit on the users' device side. When a user uploads a big pdf, that whole thing gets loaded in RAM all at once, which can slow down or even crash a tab or browser, and there's nothing we can do about that. From our testing, you’ll run into that issue before you run into our character limit.


See also the Keenious related resources in the page of this guide titled "Using with Library Databases & Other AI Tools".