Would you recommend this product?
No reviews yet
Hello everyone, super excited to be launching our 2.0 version! (https://the.iris.ai) Iris AI is your Science assistant, helping you map out relevant research for your thesis work or R&D project. She will double your productivity over existing tools such as Google Scholar. The flow is simple. Give Iris the URL to a research paper. She reads the abstract, maps out the key concepts and presents you with the most relevant articles from more than 30M Open Access papers. With a nicely visualized overview you can navigate around until you find what you need, and directly download the paper - or use the paper to make a new map. Spend your time reviewing relevant articles rather than trying to formulate key words in a field you don't know yet.  We hosted a Scithon a week ago, and learned that research teams using Iris more than double their productivity - their identification of key papers and their understanding of the problem space - compared to existing tools: http://iris.ai/journey/great-res... Also, you know, it's free for individuals. Because it's the right thing to do - science should be available to everyone. :) Happy to answer any questions, and would definitely love your feedback. (Oh, and FYI: not all papers are equally easy to read - we support at least 80% but some are still failing when you input them, please do report those so we can make sure to fix them. There are so many formats!)
Upvote (14)Share
IRIS AI is back with a new version that now understands even more content. Enter your URL and the assistant will tell you all the related research papers on the topic. Research paper example: https://the.iris.ai/map/ed76ba87... TED talk example: https://the.iris.ai/map/6728
@percival Very cool idea! I've been wanting to do something like this myself for a while. But I think it really needs to be able to better group by subject, as well as identify relevant vs. irrelevant keywords. In the example you link, half of the related papers seem to match the word "count" yielding results ranging from astrophysics to physiotherapy and AIDS(!). Hardly related to the paper in question. :) But again, definitely inspiring work! I look forward to see where this will go from here.
@askielboe @percival thank you Andreas! Introducing users the ability to spot and discard irrelevant topics is definitely in the pipeline. Although Iris is learning fast, we have not reached 100% accuracy, yet. This is why there are still some concepts in the results that are not relevant. When it comes to presenting papers across disciplines, we are actually talking about the contextual understanding that we can build with the neural networks. Thanks to that, you may find useful knowledge you didn’t expect from a field that is totally unrelated. A moment of serendipity ;)
I've always wanted an artificially intelligent virtual eye-bot to help me science! Thanks, Iris! @theirisai @percival @ezaromedia @vici4a @twitnitnit @maria_ritola
I think to do this in a meaningful way would require 2 things: H-index of the authors. Impact Factor
@datarade Impact Factor would be very useful. BTW, is there some way to calculate it?
@rmilovanov @datarade Thanks for the feedback! We are currently assessing which factors to choose when presenting the results. At the moment we are using the document relevance index. H-index and impact factor are definitely among the next candidates.
This is incredible, great job folks. Despite the paramount importance of scientific productivity, it's sad that so few groups like you are tackling it.